SQL Cheatsheet

2023-11-18T00:00:00+00:00

SQL Query Anatomy</h2>
Most databases don't like it when the SQL clauses are out of order.

Operator

`SELECT … FROM …</code> (SQLite</a>)</li>`
`INSERT INTO <table>(…) VALUES (…),…</code> (SQLite</a>)</li>`
`UPDATE <table> SET …</code> (SQLite</a>)</li>`
`DELETE FROM …</code> (SQLite</a>)</li> </ul> </li>`
`JOIN</code></li>`
`WHERE</code></li>`
`GROUP BY</code></li>`
`HAVING</code></li>`
`ORDER BY</code></li>`
`LIMIT</code></li>`
`OFFSET</code></li> </ul>`
NULL Tricks</h2>
Null in SQL is special as it isn't equal (`=</code>) to anything, not even itself, which seems wiered when coming from programming languages but makes perfect sense when considering that NULL</code> means "unknown value" wich definitely isn't equal to another "unknown value".`
`If you want null to behave more like in a programming language use the is</code> keyword, where NULL is NULL</code> evaluates to 1 opposed to NULL = NULL</code> which results in a 0.`
`Note: For all other values is</code> acts like =</code>.`
`To get a default value at the time of a query there is the coalesce()</code> function which takes multiple aruments and returns the first non-null one. (``SQLite</a>)`
`SQLite has a page with more null quirks, comparing SQLite to other Database engines.</a>`
Join Types</h2> An example SQL JOIN</code> statement.</figcaption> SELECT … FROM bookmark INNER JOIN url ON url.url_id = bookmark.url_id …</code></pre> </figure> What all joins have in common is that they glue the tables together and output the entries where the ON</code> condition is true. They differ in how they treat entries that don't make it into the output because they didn't find a "partner" using the ON</code> condition. inner</code> </dt> only returns entries that found a partner. </dd> left</code> </dt> doesn't drop columns from the already present columns, leaving the joined columns NULL</code>. </dd> right</code> </dt> additionally outputs all entries from the joined table that didn't find a partner, leaving the other columns NULL</code>. </dd> full</code> </dt> like a left</code> and right</code> join, returning all of the entries in both tables, joining the ones that are join-able and filling in where joining isn't possible. </dd> </dl> Join Examples</h3> Note: To keep this example as clean as possible types have been omitted, the SQL is intended to be compatible with sqlite3</a>. Use the command sqlite3 :memory:</code> to get an in memory database to play around with. Assume we have two tables, called left_t</code> and right_t</code> for illustration purposes, bot with a single column. CREATE TABLE left_t(lc); CREATE TABLE right_t(rc); INSERT INTO left_t VALUES ('both'),('left'); INSERT INTO right_t VALUES ('both'),('right');</code></pre> Now we can query them an see the results: SELECT lc,rc FROM left_t INNER JOIN right_t ON lc = rc ;</code></pre> We now performed an inner join of right_t</code> against left_t</code> on the single columns having the same value. The returned record is both|both</code>. Pretty intuitive. Now we substitute the INNER</code> for a LEFT</code> and the result is: both|both left|</code></pre> All rows in left_t</code> table mad it through the join, the hole in the right_t</code> was filled with a (blank) NULL</code> field. Note: If the right_t</code> table had more columns all of them would be returned as NULL</code>, independent of any DEFAULT</code> or NOT NULL</code> statements. Similar results for a RIGHT</code> join: both|both |right</code></pre> Now all rows from the right_t</code> table made it through and the missing rows in the left_t</code> table was filled with a NULL</code> entry. And for the FULL</code> join: both|both left| |right</code></pre> All rows preserved, the ones that matched got joined that ones without matches had the resulting holes filled with NULL</code> fields. Database Exploration Commands</h2> You Probably know the problem, a Database you are not familiar with with an engine you don't use every day and what were the commands again? Common Tasks and the Corresponding "SQL" Commands.</caption> Task</th> MariaDB</th> Postgres</th> SQLite</th></tr> Get Help</td> help</code></td> \h</code></td> .help</code></td> </tr> List Databases</td> show databases</code></td> \l</code></td> .databases</code></td> </tr> Use a Database</td> use <database></code></td> \c <database></code></td> Not available</td> </tr> List Tables</td> show tables</code></td> \dt</code></td> .tables</code></td> </tr> Describe Table</td> describe <table></code></td> \d <table></code></td> .schema <table></code></td> </tr> Quit Shell</td> exit</code></td> \q</code>, quit</code>, exit</code></td> .q</code></td> </tr> </table> Git Cheatsheet(s) 2023-11-03T00:00:00+00:00 Setting Up</h2> Before starting to use git one should take the few minutes to set it up. Set up your personal information that will end up on your commits: git config --global user.email <your-mail@example.org> git config --global user.name "<Name>"</code></pre> Set the default branch name to main instead of master The reasons I do this are because there was some controversy around the word "master" in this particular context and and I quickly realised that "main" is way easier to type. git config --global init.defaultBranch main</code></pre> You could set your editor for git commit messages (and other git features) using: git config --global core.editor <editor-command></code></pre> If you (like me) have one editor you want to use for everything: Set the EDITOR</code> variable in your ~/.profile</code></a> and git along with a whole range of other programs will pick up on that. Partial Adding</h2> Situation: You have made changes to one file but they should be in two different commits. git add -p [<file>]</code></pre> git will then chop up the changes into hunks and ask interactively if you want to add/stage (y</code>) a change, ignore/not stage it (n</code>), further subdivide (s</code>) or even edit (e</code>) it. You can use ?</code> to get summary of all options currently available. Undoing Things</h2> Undo last Commit</h3> Situation: You just made a git commit but you committed too much by accident. Note: If you just messed up the commit message or need to add changes use git commit --amend</code>. git reset</code> is your friend here, without any arguments it helps when you git add</code>ed too much, it makes git go into the state it was in at the last commit. git reset HEAD^1</code></pre> The HEAD^1</code></a> resets git to the parent commit of the last commit. Effectively acting as if the last commit never happened. (You can still recover it using the revlog) git reset</code> will only change your git history, not the files you have currently checked out so when using it as described above you won't loose any changes. Warning: git reset --hard</code> is intended to throw changes away and it is possible to loose uncommitted data when using it with the --hard</code> flag enabled. Restore File from last Commit</h3> Situation: You messed up a file and want to throw away the changes since the last commit. git restore <filename></code></pre> This is less destructive than git reset --hard</code> which throws away all your changes in a repository since the last commit. Shallow clone</h2> Situation: You have a very large git repository and limited time and/or bandwidth and only need the latest few commits. git clone --depth 3 <url></code></pre> This example will only fetch the last 3 commits of the main upstream branch. Shallow fetch</h3> In case you need additional branches: git fetch --depth 3 origin <branch> git switch --detach FETCH_HEAD git switch --create <branch></code></pre> After that you can use all branches you fetched as usual Other Peoples Git Guides</h2> These are guides made by other people that I think are useful. In My unorthodox, branchless git workflow</a> Drew DeVault explains how to make use of the fact that the git history of a repository isn't set in stone, I mainly use these tricks to completely avoid a lot of merge conflicts in my personal projects. Learn to change history with git rebase!</a> is Drew DeVault's cheatsheet for altering git history in general and the rebase based workflow in the blogpost liked above. In Dealing with diverged git branches</a> Julia Evans explains how to deal with the situation that changes happend on both the local and remote git repository. (She explains it better than I do here.) Ein kurzer Blick in .git/</a> is a talk held at DiVOC in 2020 by Florian Hars diving into the .git</code> directory, unfortunately no one has translated it to English yet. Inside .git</a> is a blogpost by Julia Evans that also dives into the .git</code> directory and explains what is inside. Environment Variable Zoo 2023-07-03T00:00:00+00:00 Environment variables are basically inherited, named arguments to processes with a huge impact on how a Linux or other nix systems work. Reading and Writing Environment Variables</h2> These commands here should work in almost all shells inspired by the original sh, if you use something else you probably know why and how to use it. Read out a single Variable </dt> echo "$SHELL"</code> </dd> printenv SHELL</code> </dd> Dump out all variables </dt> export</code> </dd> printenv</code> </dd> Set one or more variable for a single command </dt> TZ=UTC LANG="de_DE.UTF-8" date</code> </dd> env TZ=UTC LANG="de_DE.UTF-8" date</code> Useful program for simpler shells, environment (un)setting and chain-loading</a>. </dd> Set a variable for the currently running shell </dt> export TZ=UTC</code> </dd> declare -x TZ=UTC</code> </dd> </dl> Persisting Environment Variables</h3> For permanently setting some environment variables there are a number of options (Options further up usually can override the ones below them). .bashrc</code>, .zshrc</code>, etc. </dt> Read whenever you start your favourite shell (as a non-login shell), but don't assume it has effects on the whole session, it is not made for that. </dd> Your Desktop Environment </dt> Desktops like setting environment variables derived from their settings, sometimes for single programs, sometimes for the whole session. </dd> .profile</code> </dt> Read on every login (your login manager is responsible for doing this) and applies to your whole session. </dd> /etc/profile.d/</code> and /etc/profile</code> </dt> System-wide .profile</code> equivalent, remember that you are sharing this one with your packages. </dd> PAM </dt> PAM</a> also sets environment variables. </dd> </dl> The .bashrc</code> (plus friends) and .profile</code> can be reloaded by running source ~/<filename></code> or . ~/.filename</code>. Note: Non POSIX shells usually have their own .profile</code> equivalent, keep that in mind when changing your login shell. Bash for example uses .bash_profile</code> if it exists. Spying on other Processes</h3> Sometimes it is useful to know what environment variables a given running process has currently set, if the process belongs to you or you have root rights (use them responsibly) you have access to the file that contains the process environment. The files path is /proc/$pid/environ</code> (Substitute $pid</code> for the process id), it contains the null separated environment variables. To get them in a pretty way one can use the command tr '\0' '\n' < /proc/$pid/environ</code> (pipe it into the tr command which replaces the null separators with newlines). htop</code> shows the selected processes environment when pressing e</kbd>. Note: This is the reason why it usually is a bad idea to use environment variables for secrets, they can be read out by any other process running under the same user. When transporting secrets from one program to another use pipes and features of your shell to directly read from and write to them. Finding Documentation</h2> Outside of the usual online sources: Read the documentation and read the manpages. Most manpages have an ENVIRONMENT</code> section describing which variables the program reads and/or sets. environ(7) - The manpage about the basics of environment variables. </a> Linux and Posix</h2> You can find a more complete list of the standardised variables in the POSIX (2024, the latest at the time of writing) specification on Environment Variables</a>. HOME</code> </dt> Contains the path to your home directory, as simple as that. (Don't abuse for constructs like $HOME/.config/whatever</code>, that's what the XDG home variables are for.) </dd> PATH</code> </dt> Contains a colon separated list of paths to directories that contain binary files for running them as commands. Directories closer to the front can override others counting binaries of the same name further back. </dd> USER</code> </dt> Should contain the currently logged in username. (Fun fact: the kernel doesn't know your username) </dd> SHELL</code> </dt> Contains the path to the shell binary configured for the current user, this is probably how your favourite terminal emulator knows about your favourite shell without configuration. This may or may not be the shell you are currently using. </dd> TMPDIR</code> or TMP</code> or TEMP</code> </dt> Almost always one of them contains a path to the preferred temporary directory. Useful for when every user gets their own directory with proper permissions. TMPDIR</code> is the one in the POSIX standard and should be the preferred one. </dd> LC_</code>, LANGUAGE</code>, LANG</code> </dt> These contain locale information and are the reason why language settings work across desktop environments. Unfortunately also the reason why you have to restart applications after changing system preferences. For possible values see /etc/locale.gen</code> and /etc/locale.conf</code>. </dd> Set LANG</code> to C</code> for "no translation at all" usually that results in some form of English. </dd> Standardized in the POSIX specification on Internationalization Variables</a>. </dd> TZ</code> and TZDIR</code> </dt> These describe the current timezone to use</a> i.e. Europe/Berlin</code>. Usually points to a tzfile</a> in /usr/share/zoneinfo/</code> </dd> PAGER</code> </dt> A program that can be run using sh -c</code> that shows a textfile (unually in the current terminal) </dd> EDITOR</code> </dt> Like PAGER</code>, but for your favourite text editor. </dd> BROWSER</code> </dt> Like PAGER</code>, but for your favourite browser. (Not standardised anywhere, but a well known convention) </dd> MAILER</code> </dt> Like PAGER</code>, but for sending E-Mail using a program that is compatible with mailx</code></a>, originally intended to allow a custom implementation of /bin/mail</code> on early unixoids. Thank you to inawhilecrocodile for the reasearch!</a> (It was never standardised) </dd> Contrary to what this page previously stated it does not take URLs as argument. </dd> TERM</code> </dt> Contains the name of the terminal, together with the terminfo</a> database the capabilities of the terminal can be found out. When remoting into other machines with a less well known terminal xterm</code> is usually a sane fallback. </dd> MOTD_SHOWN</code> </dt> When set it indicates who has already shown the motd to prevent showing it twice. Convention, probably best known implementation is the pam motd module</a>. </dd> NO_COLOR</code> </dt> When set to a non-empty string Software shouldn't use ANSI-colors unless explicity configured to do so. See no-color.org</a> </dd> </dl> Note: What is meant by run with sh -c</code> is demonstrated pretty well by debians sensible-utils</a>. In a nutshell you lauch your favorite EDITOR</code> using sh -c "$EDITOR \"\$@\"" EDITOR <filepath-goes-here></code> Note: I'm trying to move the BROWSER</code> and MAILER</code> variables to the XDG section. See the related discussion on the xdg mailing list.</a> You'll have to dig around a bit to find all mails 😞. XDG freedesktop.org</h2> freedesktop.org</a> hosts quite a few specifications ensuring interoperability of free (software) desktop envoirnments, these specifications happen to contain quite a lot of envoirnment variables. XDG stands for Cross Desktop Group. Base Directories</h3> These are defines in the Desktop Base Directories Specification</a>, while the below is an overview and you should always read the specification. Note: while one can usually guess where those directories are, it is always a good idea to respect the environment variables instead of blindly writing to some guessed default path, People trying to organise their Home folder will thank you. XDG_DATA_HOME</code> </dt> Path where applications can dump their non-configuration and non-cache persistence. </dd> Default: $HOME/.local/share</code> </dd> XDG_CONFIG_HOME</code> </dt> Path for storing configuration files (and nothing else!) </dd> Default: $HOME/.config/</code> </dd> XDG_DATA_DIRS</code> </dt> colon :</code> separated list of fallback directories for XDG_DATA_HOME</code> containing default data files. (similar to how $PATH</code> works) </dd> Default: /usr/local/share/:/usr/share/</code> </dd> XDG_CONFIG_DIRS</code> </dt> colon :</code> separated list of fallback directories for XDG_CONFIG_HOME</code> containing default configuration files. (similar to how $PATH</code> works). This behaviour is very useful when administrating large desktop systems. </dd> Default: /etc/xdg/</code> </dd> XDG_CACHE_HOME</code> </dt> Path for storing cache files, may contain files specific to the current architecture details, may need cleaning up when on a shared filesystem. </dd> Also be used for files that better get lost rather than ending up in a backup by accident, keepassxc</code> is an example for that. </dd> Default: $HOME/.cache/</code> </dd> XDG_RUNTIME_DIR</code> </dt> For temporary runtime files like sockets, pipes, etc. </dd> Usually: /run/user/<userid>/</code> </dd> </dl> User Directories</h3> There is also the possibility to define User directories, these have the form of XDG_<name>_DIR</code> and are configured using the $XDG_CONFIG_HOME/user-dirs.dirs</code> and taken care of by the xdg-user-dirs-update</code> utility Commonly specified ones are: DESKTOP</code></li> DOWNLOAD</code></li> TEMPLATES</code></li> PUBLICSHARE</code></li> DOCUMENTS</code></li> MUSIC</code></li> PICTURES</code></li> VIDEOS</code></li> SCREENSHOTS</code> - Used by some popular screenshot scripts</li> </ul> Current Desktop</h3> The Freedesktop "Desktop Entry Specification"</a> specifies a XDG_CURRENT_DESKTOP</code> variable that is set from the DesktopNames</code> value in desktops session file (the ones in /usr/share/xsessions/</code> and /usr/share/wayland-sessions/</code> and whatever XDG_DATA_DIRS</code> points to). It contains a colon separated list of names for the current desktop, primarily intended for activating or deactivating desktop entries (i.e. you don't want the gnome settings cluttering up you KDE menus and vice versa). XDG Session</h3> These variables are not part of a specification, the best source for them is a guide for "Writing Display Managers" with systemd</a> and the manpage for the pam_systemd</code> module</a>. Some of them happen to be useful. XDG_SEAT</code> </dt> By definition set to seat0</code>, probably reserved for future use in multiseat scenarios. </dd> XDG_SESSION_TYPE</code> </dt> The best documentation for this one is probably the type</code> argument for the pam_systemd module, roughly it is the display server used for the session. (x11</code>,wayland</code>,mir</code>,tty</code>,unknown</code>) </dd> XDG_SESSION_CLASS</code> </dt> either user</code> or greeter</code> depending on whether the session is primarily logged in or for logging in. </dd> XDG_VTNR</code> </dt> set to the number of the TTY the current session runs in. </dd> XDG_SESSION_ID</code> </dt> Either the /proc/self/sessionid</code> of the auditing session or an "independent counter". </dd> </dl> Graphical Environment</h2> DISPLAY</code> </dt> The id of your X-Display, on your laptop it is probably :0</code>. </dd> WAYLAND_DISPLAY</code> </dt> The name of your wayland socket file in $XDG_RUNTIME_DIR/</code>. </dd> XCURSOR_SIZE</code>, XCURSOR_THEME</code>, XCURSOR_PATH</code> </dt> One way to configure how your cursor looks in most applications. It even works with wayland for when other options are not available. The Arch Wiki has more information on cursor configuration</a>. </dd> DBUS_SESSION_BUS_ADDRESS</code> </dt> Contains information on how to connect to the dbus session bus. Usually set by the dbus-run-session</code> wrapper. </dd> </dl> Mozilla</h2> These are used for Firefox, Thunderbird and other programs that are built on top of Firefox. MOZ_ENABLE_WAYLAND</code> </dt> If set it enables wayland support for all mozilla programs (if not already enabled by default) typically set to 1</code>. </dd> </dl> Java</h2> _JAVA_AWT_WM_NONREPARENTING</code> </dt> Tells java if you have a non-reparenting window manager</a>, usually set to 0 on modern Xorg, to 1 on wayland. If your window stays blank that is an indicator this one is wrong. (Yes the underscore at the start is intentional.) </dd> Possible Values: 0</code>, 1</code> </dd> </dl> More Java environment variables will be added when I run into them. Kubernetes</h2> KUBECONFIG</code> </dt> The path to a yaml file containing connection information for a Kubernetes cluster</a> that tools like kubectl</code></a> and helm</code></a> can use to connect to the cluster. </dd> k3s generates a /etc/rancher/k3s/k3s.yaml</code> file you can point this at. </dd> </dl> Fontconfig 2023-06-20T00:00:00+00:00 What is Fontconfig?</h2> Fontconfig is a library for configuring and customising font access. </blockquote> It is built into almost every program that runs on Linux and has a GUI (even in into old java software!) and makes it very easy to make configure fonts and then forget about it. You can find the Fontconfig documentation over on freedesktop.org</a>. Fontconfig CLI Tools</h2> You average Linux distribution ships Fontconfig along with some commands to poke it. fc-cache</a> </dt> for rebuilding the Fontconfig cache (after installing a font or changing configuration). Usually used like this: fc-cache -f -v</code> </dd> fc-conflist</a> </dt> gives you a list of configuration files found by Fontconfig. </dd> fc-match</a> </dt> for looking up which exact font a given name resolves to, very useful for debugging. </dd> fc-query</a> </dt> to print out metadata for a font file. </dd> fc-validate</a> </dt> can give a hint abut missing glyphs in a font file for a given language. </dd> fc-cat</a> </dt> for having a look at the Fontconfig cache. </dd> fc-list</a> </dt> to list all fonts Fontconfig knows about. </dd> fc-pattern</a> </dt> helps you by parsing Fontconfig patterns</a> , useful for debugging. </dd> fc-scan</a> </dt> similar to fc-query</code> but has the same output format as fc-pattern</code>. </dd> </dl> Font (Family) Names</h2> What one usually calls the fonts name actually is the font family name. This is because of historical reasons from back when a font was a case of lead letters for printing and one needed an extra case full of letters for the bold or italic styles. The family</code> name is usually what is accepted whenever you have some configuration file and it wants a font name. Usually the postscript name works too. Both can be found out using fc-query</code> on a given font file. fc-list</code> also shows the family name by default. Note: fc-query -b</code> omits language and character set information resulting in a more readable output. Note: Fontconfig also has a specific font name</a> that includes the family name, size, style and every other possible font setting. System Fonts</h2> System fonts using Fontconfig are just configuration aliasing generic names to your favourite font. The most common are monopace</code>, sans</code> or sans-serif</code> and serif</code>. If your desktop environment has a font preference dialog in the system settings it is changing which font names those aliases point to. Note: There is also a convention to alias emoji</code> to your preferred emoji font. Finding Fonts that Support a Codepoint</h2> To query fonts that support a given character, you need to know its codepoint. For the following examples I'll use the 😃 emoji which has the codepoint U+1f603</code>. If you have a unicode aware printf</code> command (if you don't know simply try): printf %x "'😃'"</code></pre> Which outputs 1f603</code>. There is also the possibility to use the uni</code> command. uni identify 😃</code></pre> It also lists U+1F603</code>. You could also look it up using a search engine, Wikipedia or one of those Emoji lookup sites. </blockquote> </div> You have the codepoint? Great, now you can ask the fc-list</code> command (replace the example 1f603</code> with the codepoint you are interested in.) fc-list :charset=1f603</code></pre> You should get a list of all fonts on your system that support that codepoint, if that codepoint isn't supported by a font on your system the output stays empty. Configuration Files</h2> You can find your Fontconfig configuration file in $XDG_CONFIG_DIR/fontconfig</code> which in most cases is ~/.config/fontconfig/</code>. In there you can either directly edit the fonts.conf</code> file or a file in the conf.d</code> folder. Global Configuration is in /etc/fonts/</code>. It usually consists of files linked from /usr/share/fontconfig/</code> where configuration snippets for common settings are installed. Using Predefined Configuration</h3> fc-conflist will list all configuration files for fontconfig, the active ones prefixed with +</code> the inactive ones in /usr/share/fontconfig/</code> prefixed with a -</code>. To use the predefined configuration symlink it: ln -s /usr/share/fontconfig/conf.avail/70-no-bitmaps.conf /etc/fonts/conf.d/ # Regenerate the fonconfig cache fc-cache -fv</code></pre> Note: The Void Linux Documentation on Fonts</a> gives a similar example on linking the 70-no-bitmaps.conf</code> to /etc/fonts/conf.d/</code> to mitigate the common issue that Firefox picks pixelated bitmap fonts</a> over higher quality ones. Writing Custom Configuration</h3> To make a valid Fontconfig configuration file it needs some XML scaffolding</a>: <?xml version="1.0"?> <!DOCTYPE fontconfig SYSTEM "urn:fontconfig:fonts.dtd"> <fontconfig>  </fontconfig></code></pre> That scaffold can now be filled with some rules. Note: In case you are in need of an example, have a look at the Fontconfig documentation examples</a> or my Fontconfig</a>. Those rule could be disabling fonts without having to uninstall them, useful when one doesn't have root access or wants to keep the font files for some reason. Example rule for disabling FontAwesome to prevent it from interfering with other icon-fonts.</figcaption> <selectfont> <rejectfont> <pattern> <patelt name="family"><string>FontAwesome</string></patelt> </pattern>  </rejectfont> </selectfont></code></pre> </figure> Another application is to configure the system fonts mentioned before using alias</code> rules. For setting the emoji</code> (or any other) font one simply maps the family name emoji</code> to the family name of the preferred font. An example rule for aliasing emoji</code> to Twemoji</code>.</figcaption> <alias> <family>emoji</family> <prefer> <family>Twemoji</family> </prefer> </alias></code></pre> The same mechanism could be used to alias something like Helvetica</code> to your favourite sans font family.</figcaption> </figure> The prefer</code> part of the alias also accepts multiple font families, this can be used to apply the emoji</code> alias to system fonts by telling Fontconfig to prefer your favourite monospace, sans or serif font and whatever is aliased to emoji</code> over all other installed fonts. <alias> <family>monospace</family> <prefer> <family>Noto Sans Mono</family> <family>emoji</family> </prefer> </alias></code></pre> Of course there are more applications for Fontconfig configurations, but those probably deserve more specific pages. Note: I'm using Twemoji and Noto Sans, not because I like the companies that made them, but because they are under a permissive license are mostly complete, look okay and are one of the very few things those companies are actually doing good for humanity. There are community made fonts like Inter</a> and OpenMoji</a> (plus some more) out there with an impressive coverage and feature set that shouldn't go unmentioned here. Installing fonts</h2> From packages</h3> Usually the packages do everything for you and even install some sane Fontconfig rules. If they don't, have a look at the Fontconfig files they install and check if they clash with your own configuration or if the package description, documentation or comment in its Fontconfig file mentions additional required configuration. This is usually the case for emoji fonts to get them as the default font. From files</h3> In case you have a modern Linux distribution you can install a font by dropping the ttf or otf file into ~/.local/share/fonts/</code> or ~/.fonts/</code>, the latter works on older distributions too. To apply the changes run fc-cache -f -v</code> to rebuild the Fontconfig cache and restart your applications (or log out and back in). Before restarting applications you could verify that the font was installed using the fc-match</code> and fc-list</code> commands. Converting woff2 to ttf</h4> Google (who else 🙄) made a tool for converting woff2 fonts to ttf and back</a>, the package containing it is simply called woff2</code> in most Linux distributions. It contains the woff2_decompress</code> command, give it a path to a woff2 font file and it will convert it to ttf. There is also a woff2_compress</code> that works in the other direction. ANSI Escape Sequences 2023-05-28T00:00:00+00:00 ANSI Escape sequences</a> are some standardised byte sequences that make most Terminals do some thing other than just printing characters on screen. Like printing colourful characters! Using ANSI-Escape Sequences</h2> ANSI-Escape sequences can be used by writing an ASCII-Escape characters followed by some instructions to your Programs standard output or standard error. The codes can be produced by any string escaping mechanism. In proper programming and scripting languages the syntax for the escape character in double quoted strings is usually \x1b</code>. The shell being a bit different you should use either printf</code> or echo -e</code>. While most systems support both the chance of printf</code> being supported is usually higher. Using ANSI-Escape Sequences Responsibly</h2> While colors and text effects are pretty: You should consider the cases where they shouldn't be enabled and respect user choices. Disable color output by default when the NO_COLOR</code> environment variable is non-empty</a>.</li> Disable use of escape codes by default when the output isn't a terminal.</li> Don't overuse color, it gets confusing pretty quickly.</li> Make sure no information gets lost when disabling escape codes. Does using grep on the output work to find information?</li> Does piping it into espeak-ng</code> give a sensible result?</li> </ul> </li> </ul> When writing by default</q> that means unless explicitly requested (i.e. through a --color always</code> CLI option). Seirdy about Best practices for inclusive CLIs</q>. </a> Thank you to Seirdy! Most of the above is a summarized and filtered TL;DR of Seirdys article, you should definitely read it! Setting Text Color and Effects</h2> I'll focus on Select Graphic Rendition (SGR) codes here, Wikipedia has a more complete list</a>. An example for the shell which prints out _FOOBAR_</q> in bright (01</code>) red (31</code>). After that, it resets to default settings (00</code>) and prints a newline character (\n</code>)</figcaption> printf '\033[01;31m_FOOBAR_\033[00m\n' # With %s placeholder printf '\033[01;31m%s\033[00m\n' "_FOOBAR_"</code></pre> </figure> As you can see in the example, multiple escape-codes can be chained in one sequence by delimiting them with a ;</code>. The \033</code> part will turn into an ASCII escape character, the [</code> tells the terminal that this is an ANSI Control Sequence Initiator (CSI) after that can come one or more codes. The sequence is finished by an m</code> to tell the terminal that everything between the CSI</abbr> and it are SGR</abbr> codes. Note on the \033</code> and compatibility: The \033</code> is the octal representation of the Escape ASCII-Character and is the most widely supported one across printf</code> implementations. (The dash printf implementation for example doesn't support hexadecimal representation.) Most commonly used codes</caption> Code</th> Code to reverse</th> Effect</th> </tr> 00</td> </td> Reset all effects</td></tr> Foreground color</td></tr> 30-37</td> 39</td> Set the Foreground color</td></tr> 90-97</td> 39</td> Set the Foreground color, bright variation</td></tr> Background color</td></tr> 40-47</td> 49</td> Set the Background color</td></tr> 100-107</td> 49</td> Set the Background color, bright variation</td></tr> Font effects</td></tr> 01</td> 22</td> Bright text (usually bold and brighter color)</td></tr> 02</td> 22</td> Dim text (usually only effect on color)</td></tr> 03</td> 23</td> Italic text</td></tr> 04</td> 24</td> Underlined text</td></tr> 05</td> 25</td> Blinking text</td></tr> 07</td> 27</td> Reverse foreground and background color</td></tr> 08</td> 28</td> Concealed/Hidden text</td></tr> 09</td> 29</td> Strikethrough text</td></tr> </table> Note: The leading zeros are not needed. Colors</h3> You may have noticed that the 3x</code>, 4x</code>, 9x</code>, and 10x</code> family of codes all use the last digit in the range from 0 to 7 as a color parameter. Table of colors</caption> Code</th> Color</th></tr> 0</td> Black or Dark Grey</td></tr> 1</td> Red</td></tr> 2</td> Green</td></tr> 3</td> Yellow</td></tr> 4</td> Blue</td></tr> 5</td> Violet / Purple</td></tr> 6</td> Cyan or Turquoise</td></tr> 7</td> Light Grey or White</td></tr> </table> Please keep in mind that different people have different color palettes when you want to write reusable scripts. Shell Null Termination / Separation 2023-05-27T00:00:00+00:00 Null or zero termination or separation is when one uses a null byte to separate records, which is different from the newline usually used in shell pipelines. Why</h2> Usually operating things by a newline is good enough and easy to debug, but even a little shell pipeline like a ls | sort</code> can blow up when it encounters unexpected data like a filename that contains a newline character. Fun fact: Most filesystems actually allow you to use all kinds of characters from emoji to non-printables and control characters. Including newlines. One could even put ASCII-art in a filename (I've already done that)! Using a null character saves you there as it is the only character guaranteed to never occur inside a filename and this becomes more important with other sources of (untrusted) input. The improved version would be ls --zero | sort -z</code> and a | tr '\0' '\n'</code> to convert to newlines for the human that still wants the illusion of line separation. Examples of inputs you shouldn't trust: The human in front of the machine (I know myself)</li> Your filesystem (because USB sticks, downloads, the human, etc.)</li> Anything that comes in over the network (Not even your own service, not the service you are paying for, …)</li> </ul> Translating between Worlds</h2> To translate between newline separated and the null separated world there is a little utility called tr</code></a>. null to newline </dt> tr '\0' '\n'</code> </dd> newline to null </dt> tr '\n' '\0'</code> </dd> swap newline and null </dt> tr '\n\0' '\0\n'</code> </dd> remove potential nulls </dt> tr -d '\0'</code> </dd> </dl> Make sure to actually remove the characters you assume to not be in your input! Getting a Program into Null-Separation Mode</h2> Unfortunately there is no "the one way" to make a program use nulls instead of newlines so here is a a hopefully useful table. Please note that some programs only have the option available in the gnu version, but not in i.e. the busybox version. If a program has some kind of printf</code> option one can use that to make the output null separated. Command</th> Version</th> Input Flag(s)</th> Output Flag(s)</th> </tr> awk</td> gnu, other modern</td> -v 'RS=\0'</td> </td></tr> cut</td> gnu</td> -z</td> -z</td></tr> find</td> </td> </td> -print0</td></tr> fzf</td> </td> --read0</td> --print0</td></tr> grep -r</td> gnu</td> </td> -Z</td></tr> grep</td> gnu</td> -z</td> -z</td></tr> inotifywait</td> </td> </td> --format '…%0' --no-newline</td></tr> ls</td> gnu</td> </td> --zero</td></tr> rg</td> </td> </td> -0</td></tr> sed</td> gnu</td> -z</td> -z</td></tr> sort</td> gnu, busybox</td> -z</td> -z</td></tr> tar -T</td> gnu</td> --null</td> </td></tr> uniq</td> gnu, busybox</td> -z</td> -z</td></tr> xargs</td> gnu, busybox</td> -0</td> </td></tr> read</td> </td> IFS="" read</td> </td></tr> </table> Note: The best way to automatically find out if a program supports an option is to grep -q</code> across the output of its --help</code>. Just make sure to choose a specific enough regex to avoid false positives. Trying out is also an option but that usually is a bit more complex. Note: read</code> is a shell builtin, which is why you don't have to export the IFS</code> (information separator) variable. This also affects other shell builtins like for-loops. Shell Variables and Null Bytes</h2> Shells usually drop null-characters when storing them into variables, so don't rely on that happening and pipe information directly or, if your constraints allow</a>, use bash arrays</a>. Because of this limitation for-loops can't make use of null separation, use while read</code> (will probably result in a subshell though) and xargs</code> instead. Note on IFS</code>: Setting the IFS</code> to a newline by IFS="<newline>"</code> where <newline></code> is a literal line break (as in: the closing quote being the first character of the next line) you at least get clean newline separation if a for loop is really needed. Speech Recognition Frontends 2023-05-09T00:00:00+00:00 This page intends to give an overview what kind of projects out there are open source (or preferably free software) making use of speech recognition without relying on some well known cloud providers. Overview</h2> The projects listed here fall into 3 broader categories. Build your own voice something: voice2json</a></li> Dragonfly</a></li> </ul> Voice Input Methods that could replace Mouse and Keyboard: EasySpeak</a></li> Numen</a></li> Simon</a></li> </ul> Voice assistants: Rhasspy</a></li> Mycroft</a></li> </ul> voice2json</h2> voice2json</code> is a collection of command-line tools</a> for offline speech/intent recognition on Linux. It is free, open source (MIT</a>), and supports 18 human languages</a>. </blockquote> https://voice2json.org/ </a> It can make use of CMU Sphinx</a>, Kaldi</a>, Mozilla DeepSpeech</a> and Julius</a>, depending on the used language profile. Dragonfly</h2> Dragonfly is a speech recognition framework for Python that makes it convenient to create custom commands to use with speech recognition software. </blockquote> https://github.com/dictation-toolbox/dragonfly </a> I learned about this from handsfree.dev</a>. EasySpeak</h2> Voice control for Linux desktops. Fully local, no cloud, Wayland-native. </blockquote> https://github.com/ctsdownloads/easyspeak </a> Tries to be a blend of input method and voice assistant. Numen</h2> Numen is voice control for computing without a keyboard. It works system-wide on Linux and the speech recognition runs locally. </blockquote> https://numenvoice.org/ </a> John Gebbie</a> also has some related projects linked in the README</a>, he also made handsfree.dev</a>. Simon (KDE)</h2> Simon is an open source speech recognition program that can replace your mouse and keyboard. The system is designed to be as flexible as possible and will work with any language or dialect. Simon uses the KDE libraries</a>, CMU SPHINX</a> and / or Julius</a> coupled with the HTK</a> and runs on Windows and Linux. </blockquote> https://simon.kde.org/ </a> Note:Some links have been replaced with their redirect targets. Rhasspy Voice Assistant</h2> Rhasspy (ɹˈæspi) is an open source</a>, fully offline set of voice assistant services</a> for many human languages […] </blockquote> https://rhasspy.readthedocs.io </a> Mycroft Voice Assistant</h2> Mycroft is an open source voice Assistant that makes use of a custom wakeword listener</a> and Mozilla DeepSpeech</a> together with some custom components that has no easy-to-find non-marketing short description. If you are looking for the technical stuff instead of company selling things stuff, you'll find it in the Documentation menu of their page. https://mycroft.ai/ </a> Feedback</h2> If some of the information here is outdated, incorrect or you know a project that is not in this list please open an issue over on Codeberg</a> or contact me in some other way</a>. Scanning, OCR and PDF 2023-04-24T00:00:00+00:00 Scanning</h2> Under Linux one can scan an image with the scanimage</code> command provided by the SANE framework</a>. A helper function that scans an image from the default scanner to the given path with settings suitable for OCR</figcaption> scan_image_to() { printf "Scanning Image …\n" scanimage --mode Gray --format png --resolution 300 -o "$1" -p }</code></pre> Note: The --mode</code> and --resolution</code> options are scanner specific and my work different for you. See scanimage --help</code>. </figure> Unpaper</h3> Unpaper</a> (manpage</a>) is a scan post-processing tool which can remove paper and scan artefacts and also do alignment of carelessly scanned pages. Which is great when preparing them for OCR. Thanks to a friend of mine for mentioning it! OCR using Tesseract and imagemagick</h2> Tesseract</a> is both, a library and command line tool</a> for Optical Character Recognition (OCR). Image in, text in image out. It best works with greyscale images with low background noise and high contrast. Getting a scan closer to that can be done using unpaper or - for scans of known quality - imagemagick using its builtin contrast filter and the level clipping, which is faster than any fully automatic tool. A function that can be used to OCR a given prerotated image that does some preprocessing with imagemagick to improve noise (the -level option) and contrast for scans, the output is the OCR as plaintext to stdout.</figcaption> # Usage: ocr_image <image> ocr_image() { convert "$1" -colorspace Gray -contrast -level 20%,100% -contrast -contrast - | \ tesseract -l deu+eng --oem 2 --psm 3 - - }</code></pre> </figure> To explain the most important tesseract options: -l </dt> To set the used language or script</a>. Accepts multiple values separated by a +</code>. i.e. -l deu+eng</code> </dd> --oem </dt> To set the mode of operation, meaning whether or not to use the LSTM</a> based approach or the original tesseract OCR. At the time of writing I had the best results with option 2, enabling both. </dd> --psm </dt> Tells tesseract what kind of layout analysis</a> to run. I prefer to do a separate OSD (Orientation and Script Detection) only run in mode 0 rotating the image and then using the full automatic layout mode 3 without OSD to do the actual OCR. </dd> --user-words </dt> To give tesseract a text file with a list of words that are likely to occur in the document, this helps avoiding the OCR equivalent to a typo. There is also a pattern file (note that these are not regexes!). More on word and pattern lists in the tesseract manpage.</a> </dd> </dl> Note for Void-Linux users: The package name and command for calling tesseract is tesseract-ocr</code> on Void. This function uses tesseract with the psm 0 mode to correctly orientate a given image file that is assumed to be German or English text. The file is then replaced.</figcaption> # Usage: rotate_scanned_image <image> rotate_scanned_image() { # Replace or leve out the -l option here if you work with other languages/scripts PAGE_OSD="$(tesseract -l deu+eng --oem 2 --psm 0 "$1" - 2>&1)" #if DEBUG_OSD is set print some debug output. [ -n "$DEBUG_OSD" ] && printf "PAGE OSD =====\n%s\n==============\n" "$PAGE_OSD" >&2 APPLY_ROTATION="$( printf "%s\n" "$PAGE_OSD" | awk '/^Rotate:/{ print $2 }' )" convert "$1" -rotate "${APPLY_ROTATION:-0}" "$1" }</code></pre> </figure> Note on DPI settings: Tesseract assumes the images to be scanned at 300dpi, if you are using a different resolution make sure to tell tesseract using the --dpi</code> option. PDF creation and manipulation</h2> Creating a PDF from images</h3> To create a PDF from images without OCR one can use imagemagick. Example that uses the convert</code> command to convert two png files to an pdf with a pagesize of DIN A4.</figcaption> convert -page a4 page_1.png page_2.png document.pdf</code></pre> </figure> Also see the imagemagick documentation on the -page option.</a>. OCR using OCRmyPDF</h3> OCRmyPDF</a> is a python wrapper for tesseract that makes it pretty painless to run OCR on a PDF and turn PDFs into standards compliant PDF/A</a> and optimize PDFs to reduce filesize. An example for OCRing a PDF file that contains German and English text as images.</figcaption> ocrmypdf -l deu+eng input.pdf output.pdf</code></pre> </figure> Package Note: The package is called ocrmypdf</code> for all major Linux distributions (AUR on Arch). Thank you (again) to the friend who found this tool. Creating PDFs with OCR using tesseract and poppler</h3> To create PDFs with an OCR overlay one can delegate the image to PDF step to tesseract using its pdf</code> preset and join the resulting documents after the OCR step with the poppler pdfunite</code></a> tool. Tesseract command to generate a PDF with OCR overlay for a single page with the options left out (they are the same as with the plaintext preset)</figcaption> tesseract page_n.png page_n [options...] pdf</code></pre> </figure> pdfunite command to join multiple PDFs together. Btw. these can be any PDF documents.</figcaption> pdfunite page_.pdf document.pdf</code></pre> </figure> Accessibility Note: These OCR overlays while pretty good are intended for making the documents searchable, they are not a replacement for proper semantic and accessible documents. Getting text out of PDF files</h3> To get text out of a PDF one can use the pdftotext</code></a> and pdftohtml</code></a> command (which come with poppler). pdftotext file.pdf</code></pre>Packages</h2> To get the commandline utilities on some distributions one should also install the <packagename>-bin</code> package. Imagemagick is sometimes just called magick</code>. Tesseract usually comes without the needed data files, these probably come in packages called like tesseract-data-<lang></code>, if you just want everything there is probably a tesseract-data-all</code> metapackage. The pdfunite</code> and pdfto</code> commands are either directly in the poppler</code> package or in poppler-utils</code>. IP-Address Ranges 2023-03-08T00:00:00+00:00 In General</h2> While IPv4 and IPv6 have different terminology, they share some concepts. Scope and Delivery schemes. Both IP-Versions share the following delivery schemes: Unicast </dt> one to one, the default </dd> Multicast </dt> one to many, v6 had a huge Update there </dd> Broadcast </dt> usually inside the scope of a single network </dd> </dl> Scope is a bit version dependant, usually you have: Global </dt> These addresses are globally routable and assigned by the IANA</a> and the local Internet registries. </dd> Private </dt> These addresses are reserved for private use and you are free to use the in your own network, just be aware, that you can't use them on the internet. </dd> Loopback </dt> These addresses always route back to the sending machine, usually localhost</code> resolves to them. </dd> Link-Local </dt> These adresses are automatically assigned so that host on the same network can talk to each other. </dd> Documentation </dt> These addresses are reserved for documentation purposes. </dd> Other reserved </dt> These adresses are reserved and have some conditions on when you are supposed to use them. </dd> </dl> IPv6 Adress Ranges</h2> By default IPv6 adresses are unassigned global unicast adresses and therefore reserved for future use. The IANA has a complete list with the IANA IPv6 Special-Purpose Address Registry</a>. Global Unicast (RFC 4291, Section 2.5.4</a>) </dt> 2000::/3</code> </dd> Unique Local (Private) (RFC 4193</a>) </dt> fd00::/8</code> (with Random Global ID) </dd> fc00::/8</code> (reserved for future use) </dd> Loopback (RFC 4291, Section 2.5.3</a>) </dt> ::1/128</code> </dd> Link-Local (RFC 4291, Section 2.5.6</a> </dt> fe80::/10</code> </dd> Documentation (RFC 3849</a>) </dt> 2001:db8::/32</code> </dd> Multicast (RFC 4291, Section 2.7</a>) </dt> ff00::/8</code> </dd> IETF Protocol Assignments (RFC 6890</a>) </dt> 2001:0000::/23</code> </dd> NAT64 Well-Known Prefix (RFC 6052, Section 2</a>) </dt> 64:ff9b::/96</code> </dd> </dl> Note: Site-Local IPv6 Unicast Addresses (RFC 4291, Section 2.5.7</a>) have been deprecated in favour of Unique Local Addresses (RFC 4193</a>) which are pretty similar, but better defined. IPv4 Address Ranges</h2> By default IPv4 adresses are adresses that fall in the global unicast category. The IANA has a complete List with the IANA IPv4 Special-Purpose Address Registry</a>. Private (RFC 1918</a>) </dt> 192.168.0.0/16</code> </dd> 172.16.0.0/12</code> </dd> 10.0.0.0/8</code> </dd> Loopback (RFC 1122, Section 3.2.1.3</a>) </dt> 127.0.0.0/8</code> Note that this is an entire network. </dd> 127.0.0.1</code> is usually used. </dd> Link-Local (RFC 2927</a>) </dt> 169.254.0.0/16</code> </dd> Documentation (RFC 5737</a>) </dt> 192.0.2.0/24</code> (TEST-NET-1) </dd> 198.51.100.0/24</code> (TEST-NET-2) </dd> 203.0.113.0/24</code> (TEST-NET-3) </dd> Shared (RFC 6598</a>) </dt> 100.64.0.0/10</code> used as Shared Address Space to accommodate the needs of Carrier-Grade NAT (CGN) devices.</q> </dd> Benchmarking (RFC 2544</a>) </dt> 198.18.0.0/15</code> </dd> Multicast </dt> 240.0.0.0/4</code> Multicast "Namespace" (RFC 1112</a>) </dd> 192.52.193.0/24</code> Automatic Multicast Tunneling (RFC 7450</a>) </dd> IETF Protocol Assignments (RFC 6890</a>) </dt> 192.0.0.0/24</code> </dd> </dl> Broadcast</h3> Broadcast addresses with IPv4 work by setting the host-part of the address to all ones (in the binary representation) they send to all hosts in the local network. Note: In RFC 919</a> 255.255.255.255</code> is defined as broadcast to all of its immediate neighbors</q>. Serial Port Troubleshooting on Linux 2023-02-14T00:00:00+00:00 A serial port (i.e. for programming a microcontroller) can have multiple reasons why it is not working. This guide is based on my experience and will try to cover most cases, FAQ style. Hints</h2> It is always a good idea to try to look up the problem in your distributions wiki/documentation first! You can manually connect to a serial port using screen /dev/tty<name></code>. Quit with Ctrl+a</kbd> and pressing k</kbd> after that. Is your Serial device showing up?</h2> Use ls /dev/tty</code> to get a list of terminal devices on your Linux machine. If you find a ttyUSBn</code> or a ttyACMn</code> (where n</code> is a number) that appears when you plug your device in and disappears when unplugging (yes that's the easiest way) that is your device with hardware and drivers working, skip this section. Note: Builtin serial ports are sometimes also called /dev/Sn</code>. Is the Hardware recognised?</h3> Try running lsusb</code>, it will give you a list of connected USB devices, if your device is not among them you either have a bad cable, a bad microcontroller or (for some micros) the controller is not in programming mode. If the device shows up in lsusb</code> but there is no serial port, your hardware is working but there is a driver problem. (or the hardware doesn't provide a serial port.) Would a reboot fix the problem?</h3> This only the case if there has been a system update since the last reboot. (Depending on your definition of a shortcut: just reboot) If you are not using a very uncommon setup check if /lib/modules</code> (the directory kernel modules (drivers) are in) contains a directory called the same as the output of uname -r</code> (the version of the currently running kernel). If that is not the case your package manager has cleaned up the old modules during an update and you have not rebooted yet (still running the old kernel). The solution in this case is a reboot, then the hardware management will be able to find the modules the kernel needs because the running version matches the version of the modules in the filesystem. Missing Package or Driver Error</h3> If a reboot did not or would not fix your problem try searching for a driver package (usually the package name contains ftdi</code> for USB to serial adaptors). If you find the necessary package already installed, running dmesg -w</code> as root and then plugging in the device is also worth a look at this point. dmesg</code> prints logs from your kernel, including potential information and errors from relevant drivers. Do you have permission for accessing serial ports?</h2> Run ls -l /dev/tty</code>, this will tell you who owns the device files, file permissions apply like with regular files. An example output line, the file belongs to the root</code> user and the dialout</code> group, both are allowed to read and write.</figcaption> crw-rw---- 1 root dialout 4, 67 5. Sep 12:25 /dev/ttyS0 #^ ^ The 2 rw here indicate that owner and group may read from and write to this device.</code></pre> </figure> By default on many distributions the hardware management comes preconfigured in a way that only permits users in the dialout</code> group to use serial ports. Check if you are a member of that group using the groups</code> command. Note: The dialout</code> group may have a different name on some distributions if the ls -l</code> didn't indicate that the serial port belongs to the dialout</code> group this is the case for you. (look up serial ports in the distributions wiki). Note on Arch derived Distributions: On Archlinux and derived distributions the relevant group is called uucp</code> instead of dialout</code>. (UUCP stands for Unix-to-Unix Copy</a> which was common to be used over a serial link.) You can give a user permission by running sudo usermod -a -G dialout <username></code> and logging out and back in after that. In case your system looks completely different you may want to investigate udev</code> and udev rules</a>. Is something else using the port?</h2> Since a program doesn't really know what is on the other side of a serial port some programs simply connect and try to find out. This may interfere with whatever you are trying to do. One of these programs is brltty</a> which is preconfigured by most big distributions. It tries very hard to make sure plugging a braille display into an USB port will make it immediately available. (which is very good, your little Arduino isn't nearly as important!) Unfortunately it has to open every serial device that connects to the system to find out whether it is a braille display or not. To fix that behaviour either configure the "offending" program or stop (and disable) the service it belongs to if you don't need it. Example of how to disable and stop the brltty</code> service on most systemd based distributions. Do not do this if you rely on a working braille display!</figcaption> sudo systemctl disable --now brltty</code></pre> </figure> Did that fix it?</h2> If this guide helped you please consider sharing it. In case you found a mistake or ended up in a dead end, don't hesitate to contact me</a>. arduino-cli 2023-02-13T00:00:00+00:00 While arduino-cli</code> is an awesome tool, unfortunately most guides skip right past it and assume you use the Arduino IDE, this page is a collection of common patterns for arduino-cli</code>. It assumes that you know a few things about how Arduino works. Basics</h2> To create a sketch arduino-cli</code> assumes that your main .ino</code> file has the same name as the directory containing it. To build and upload use arduino-cli compile --fqbn <fqbn></code> and arduino-cli upload -p <serial-device> --fqbn <fqbn></code> in your code directory. Hint: On Linux your serial devices are called /dev/ttyUSBn</code> and /dev/ttyACMn</code> where n</code> is a number. Make sure you have the permissions to access them and nothing else is trying to use those ports.</a> You've got an unhelpful error message? Try adding the --log</code> flag, it will cause arduino-cli</code> to print out whatever it is currently doing. This way one can get some context for the error. FQBN? Fully Qualified Board Name!</h2> The Fully qualified Board name consists of what Arduino calls the Platform ID and a name for the board. You can find it out either by using arduino-cli board search <searchterm></code>, arduino-cli board listall</code> or by appending the official board name in lowercase to the Platform ID. Examples include: arduino:avr:uno</code> (Arduino Uno or another board with an ATMega328P running at 16MHz)</li> esp8266:esp8266:generic</code> (A generic esp8266 Module)</li> esp32:esp32:esp32</code> (A generic esp32 module)</li> </ul> Please note that the esp boards are not available by default. Compile and Upload a Sketch using arduino-cli</h2> I usually create a compile-and-upload.sh</code> script to make development and testing easier. #!/bin/sh arduino-cli compile --fqbn arduino:avr:uno && \ arduino-cli upload --fqbn arduino:avr:uno -p "${1:-/dev/ttyUSB0}"</code></pre> A simple compile and upload script with a fixed board and an optional argument for the serial port that defaults to /dev/ttyUSB0</code></figcaption> </figure> Note: When given the -u</code> flag and the options needed for uploading arduino-cli compile --fqbn arduino:avr:uno -u -p "${1:-/dev/ttyUSB0}"</code> behaves like the above example script. Note on shell syntax: The "${1:-/dev/ttyUSB0}"</code> is read by your shell as replace this with the first argument given to the script or with /dev/ttyUSB0</code>. Adding more Boards</h2> The arduino-cli</code> is not just a wrapper for a compiler and whatever you prefer for uploading … it also is a package manager. If you want more boards (or libraries) you can search for and install them. But if you want to write code for a non-Arduino branded board (i.e. an ESP8266 or ESP32) those packages are not in the Arduino package repositories. Adding an extra repository can be done by letting arduino-cli</code> do the work … arduino-cli config add board_manager.additional_urls 'url://to/index.json'</code></pre> … or by manually editing ~/.arduino15/arduino-cli.yaml</code> and adding the URL pinting to the respositories _index.json</code> file there. board_manager: additional_urls: - https://arduino.esp8266.com/stable/package_esp8266com_index.json - https://dl.espressif.com/dl/package_esp32_index.json - https://raw.githubusercontent.com/espressif/arduino-esp32/gh-pages/package_esp32_dev_index.json</code></pre> An excerpt from the configuration file from my laptop with additional URL to the stable esp8266 and esp32 repositories in addition to the development version of the esp32 repository.</figcaption> </figure> Don't forget to arduino-cli update</code> and arduino-cli upgrade</code>! Flashing a Bootloader using arduino-cli</h2> arduino-cli</code> already knows how to flash a new bootloader with the burn-bootloader</code> subcommand, you have to tell it for which board you want the bootloader using --fqbn</code> and which programmer you are using with -P</code>. Note: Programmers usually aren't quite cheap, but if you have another Arduino (compatible) with UART (serial) and SPI interface you can use that Arduino as an ISP</abbr> </a>. (use -P arduinoasisp</code>) It may be necessary to provide the port the programmer is on via the -p</code> option. (like with the upload</code> command) Example using the official AVRISP mkII programmer for AVR chips for an ATMEL 328P in an Arduino Uno like configuration</figcaption> arduino-cli burn-bootloader --fqbn arduino:avr:uno -P avrispmkii</code></pre> </figure> To find out which programmer string you need, go to the ArduinoCore-</code> repository of your favourite architecture (See github.com/arduino</a> for the official ones) and open the programmers.txt</code> file (i.e. ArduinoCore-avr/programmers.txt</a>), the keys used there are the needed ids for the -P</code> option. Note: An interesting option might be to use an Arduino Microcontroller as your programmer</a>, use -P arduinoisp</code> in this case. The neccessary code is also in a public repository.</a> Why do I need a Bootloader?</h3> To be able to seamlessly upload code with just a USB to serial converter like in the example above your microcontroller needs to have the Arduino bootloader installed, a small piece of software that runs when the controller starts (that's the reason you need to hook up the reset line when programming) that can be told to write a new program to the controller instead of starting the already existing one. The boards you can buy online come with a bootloader preinstalled. But maybe you want to build your own Arduino compatible or released the magic smoke and want to just buy the chip itself without a development-board, that is when you want to flash the bootloader yourself. Do I need a Bootloader to use the Arduino Framwork?</h3> No, you can use the programmer of your choice (just use the -P</code> option with the upload command) to flash whatever you coded up directly into the controller at the cost of having to use the programmer every time you upload a new program. (Unless that program is a bootloader 😉) Using sketch-local Libraries</h2> To use liberies stored anywhere but the default location (i.e. in a libs</code> folder next to your .ino</code> files) the --library</code> and --libraries</code> can be used. Both accept a comma seperated list of filepaths. --library</code> points to individual libraries and --libraries</code> points to folders of libraries. Example to compile a sketch with a libs</code> folder</figcaption> arduino-cli compile --libraries libs/ --fqbn <fqbn></code></pre> </figure> You may also be interested in</h2> The official arduino-cli documantation </a> Serial Port Troublehooting on Linux </a> Resistor Value Table 2023-01-14T00:00:00+00:00 How To Read Resistors</h2> Color</th> Value</th> Multiplier</th> Tolerance</th></tr> Black</td> 0</td> 1Ω</td> </td></tr> Brown</td> 1</td> 10Ω</td> ± 1%</td></tr> Red</td> 2</td> 100Ω</td> ± 2%</td></tr> Orange</td> 3</td> 1KΩ</td> </td></tr> Yellow</td> 4</td> 10KΩ</td> </td></tr> Green</td> 5</td> 100KΩ</td> ± 0.5%</td></tr> Blue</td> 6</td> 1MΩ</td> ± 0.25%</td></tr> Violet</td> 7</td> 10MΩ</td> ± 0.10%</td></tr> Gray</td> 8</td> </td> ± 0.05%</td></tr> White</td> 9</td> </td> </td></tr> Gold</td> </td> 0.1Ω</td> ± 5%</td></tr> Silver</td> </td> 0.01Ω</td> ± 10%</td></tr> </tbody> </table> Resistors are usually marked with 4 or 5 colored rings to indicate value and tolerance. To read the code in the right direction you have to know that the order is: 1st Ring, 2nd Ring, 3rd Ring (only when you have a 5-Ring code), Multiplier Ring, a larger gap (sometimes not), Tolerance Ring To decode, read the 1st to 3rd Ring as if they were decimals and then multiply that number with the Multiplier. For SMD resistors the pattern apparently repeats itself where the last digit on the package is a Multiplier for the other digits, resulting in a value in ohms. Also the form where there is an R</code> in place of the decimal point is common for smaller values. Podcasts 2022-11-06T00:00:00+00:00 Note that these are mostly about IT and (I think) well known podcasts. English Language Podcasts</h2> FOSS and Crafts</h3> https://fossandcrafts.org/ </a> FOSS and Crafts is an interdisciplinary exploration of collaborative creation. […] We explore such topics as computer programming, craft production, user freedom and agency, especially informed by the principles of the F(L)OSS (Free, Libre, and Open Source Software) and free culture movements. </blockquote> postmarketOS Podcast</h3> https://cast.postmarketos.org/ </a> You just found the corner of the Internet, where postmarketOS team members discuss whatever crazy development details we encounter while cooking up our favorite mobile Linux distribution. Besides news and anecdotes, once in a while we throw in an interview with an exciting guest who brings us their unique view on the wider mobile Linux scene. </blockquote> Deutschsprachige Podcasts</h2> Chaosradio</h3> https://chaosradio.de/ </a> Chaosradio ist der Podcast des Chaos Computer Club Berlin (CCCB). In dem (bis zu) zweistündigen Format informieren wir seit 1995 über technische und gesellschaftliche Themen. Weiterlesen …</a> </blockquote> C-RadaR</h3> https://www.c-radar.de/ </a> Monatliche Radiosendung des Chaos Computer Clubs auf Radio Darmstadt. Jeden 2ten Donnerstag im Monat, 21-23 Uhr. 103,4 MHz UKW / DAB+ / Stream. Tune In! </blockquote> Pentaradio</h3> https://www.c3d2.de/radio.html </a> Magazin für Netzkultur und Politik des Chaos Computer Clubs Dresden C3D2. jeden vierten Dienstag im Monat 21:30 – 23:00 Uhr (Beschreibung: coloRadio</a>)</cite> </blockquote> Webdirectories 2022-11-03T00:00:00+00:00 Webrings</h2> Listed roughly in order of preference, it is NOT a ranking. Prefernce in this case means: In therms of how many sites I personally conider interesting. Webrings I found more recently or just don't explore as often will naturally rank lower. In case you are new to webrings: Webrings are sometimes members of other webrings and a few sites are members of a lot of webrings so the below are just some possible entry points. Geekring</h3> https://geekring.net </a> For all those funny, quirky, interesting, personal, strange, offbeat websites that's not page 1 on the search engines. </blockquote> Hotline Webring</h3> Has no restrictions for joining and may contain objectionable content. https://hotlinewebring.club/ </a> Geekring with more retro themed websites. The retronaut webring</h3> Has no restrictions for joining and may contain objectionable content. https://webring.dinhe.net/ </a> A webring that contains all kinds of sites. The Yesterweb Ring</h3> https://yesterweb.org/webring/ </a> This webring is for anyone who is tired of how boring and same-y the internet is today. It's for anyone who is sick of seeing websites used purely to drive monetization, informative blogs that ask you to subscribe to see content. </blockquote> XXIIVV Webring</h3> https://webring.xxiivv.com/ </a> This webring is an attempt to inspire artists & developers to build their websites and share traffic amongst each other. The ring welcomes hand-crafted wikis and portfolios. </blockquote> The 512KB Club</h3> https://512kb.club/ </a> The 512KB Club is a collection of performance-focused web pages from across the Internet. </blockquote> More of a directory but sometimes considered a webring (it's missin the ring-part though) envs - webring</h3> https://envs.net/ring/ </a> this webring can be joined by any user on envs.net. </blockquote> Actual web directories</h2> The real phonebooks of the internet! ooh.directory</h3> https://ooh.directory/ </a> A collection of 2,295 blogs about every topic </blockquote> The counter in the description is probably outdated by the time you are reding this 😃. Sites with lots of interesting links</h2> Seirdy’s Home</h3> https://seirdy.one/ </a> Not only a source of great links but also great articles containing them. TinyGem</h3> https://tinygem.org/ </a> Have you ever bookmarked a page and worried that you will not find the same stuff next time you are there? Keep the pages on the web saved forever. </blockquote> A searchable link aggregator for links people find interesting. ASCII Table 2022-10-16T00:00:00+00:00 This ASCII-Table is mostly derived from the Unicode 15.0 Documentation</a>. Mostly Printable Characters</h2> Skip to the control characters table </a> Codes 0x20 to 0x7E are referred to as the printable characters, 0x7F is a control character. Wikipedia has more background information.</a> Numbers and Specials</caption> Char</th> Hex</th> Dec</th></tr> SPACE</td> 20</td> 32</td></tr> !</td> 21</td> 33</td></tr> "</td> 22</td> 34</td></tr> #</td> 23</td> 35</td></tr> $</td> 24</td> 36</td></tr> %</td> 25</td> 37</td></tr> &</td> 26</td> 38</td></tr> '</td> 27</td> 39</td></tr> (</td> 28</td> 40</td></tr> )</td> 29</td> 41</td></tr> </td> 2a</td> 42</td></tr> +</td> 2b</td> 43</td></tr> ,</td> 2c</td> 44</td></tr> -</td> 2d</td> 45</td></tr> .</td> 2e</td> 46</td></tr> /</td> 2f</td> 47</td></tr> 0</td> 30</td> 48</td></tr> 1</td> 31</td> 49</td></tr> 2</td> 32</td> 50</td></tr> 3</td> 33</td> 51</td></tr> 4</td> 34</td> 52</td></tr> 5</td> 35</td> 53</td></tr> 6</td> 36</td> 54</td></tr> 7</td> 37</td> 55</td></tr> 8</td> 38</td> 56</td></tr> 9</td> 39</td> 57</td></tr> :</td> 3a</td> 58</td></tr> ;</td> 3b</td> 59</td></tr> <</td> 3c</td> 60</td></tr> =</td> 3d</td> 61</td></tr> ></td> 3e</td> 62</td></tr> ?</td> 3f</td> 63</td></tr> </table> Uppercase</caption> Char</th> Hex</th> Dec</th></tr> @</td> 40</td> 64</td></tr> A</td> 41</td> 65</td></tr> B</td> 42</td> 66</td></tr> C</td> 43</td> 67</td></tr> D</td> 44</td> 68</td></tr> E</td> 45</td> 69</td></tr> F</td> 46</td> 70</td></tr> G</td> 47</td> 71</td></tr> H</td> 48</td> 72</td></tr> I</td> 49</td> 73</td></tr> J</td> 4a</td> 74</td></tr> K</td> 4b</td> 75</td></tr> L</td> 4c</td> 76</td></tr> M</td> 4d</td> 77</td></tr> N</td> 4e</td> 78</td></tr> O</td> 4f</td> 79</td></tr> P</td> 50</td> 80</td></tr> Q</td> 51</td> 81</td></tr> R</td> 52</td> 82</td></tr> S</td> 53</td> 83</td></tr> T</td> 54</td> 84</td></tr> U</td> 55</td> 85</td></tr> V</td> 56</td> 86</td></tr> W</td> 57</td> 87</td></tr> X</td> 58</td> 88</td></tr> Y</td> 59</td> 89</td></tr> Z</td> 5a</td> 90</td></tr> [</td> 5b</td> 91</td></tr> \</td> 5c</td> 92</td></tr> ]</td> 5d</td> 93</td></tr> ^</td> 5e</td> 94</td></tr> _</td> 5f</td> 95</td></tr> </table> Lowercase</caption> Char</th> Hex</th> Dec</th></tr> `</td> 60</td> 96</td></tr> a</td> 61</td> 97</td></tr> b</td> 62</td> 98</td></tr> c</td> 63</td> 99</td></tr> d</td> 64</td> 100</td></tr> e</td> 65</td> 101</td></tr> f</td> 66</td> 102</td></tr> g</td> 67</td> 103</td></tr> h</td> 68</td> 104</td></tr> i</td> 69</td> 105</td></tr> j</td> 6a</td> 106</td></tr> k</td> 6b</td> 107</td></tr> l</td> 6c</td> 108</td></tr> m</td> 6d</td> 109</td></tr> n</td> 6e</td> 110</td></tr> o</td> 6f</td> 111</td></tr> p</td> 70</td> 112</td></tr> q</td> 71</td> 113</td></tr> r</td> 72</td> 114</td></tr> s</td> 73</td> 115</td></tr> t</td> 74</td> 116</td></tr> u</td> 75</td> 117</td></tr> v</td> 76</td> 118</td></tr> w</td> 77</td> 119</td></tr> x</td> 78</td> 120</td></tr> y</td> 79</td> 121</td></tr> z</td> 7a</td> 122</td></tr> {</td> 7b</td> 123</td></tr> |</td> 7c</td> 124</td></tr> }</td> 7d</td> 125</td></tr> ~</td> 7e</td> 126</td></tr> DEL</td> 7f</td> 127</td></tr> </table> </div> Note on the DEL</code> code:The DEL being 0x7F, meaning it has all 7 bits set, probably originates from how one "deleted" something from paper tape, by just punching all of the holes one could "erase" the original character. Its meaning is ambigious, but it usually refers to a backspace. Fun Fact: The lowercase characters in ASCII were an "aftertought" and only added later in 1965, two years after the initial release. Control characters</h2> Char</th> Hex</th> Oct</th> Dec</th> Name</th> C Esc</th></tr> NUL</td> 00</td> 000</td> 0</td> Null</td></tr> SOH</td> 01</td> 001</td> 1</td> Start Of Heading</td></tr> STX</td> 02</td> 002</td> 2</td> Start Of Text</td></tr> ETX</td> 03</td> 003</td> 3</td> End Of Text</td></tr> EOT</td> 04</td> 004</td> 4</td> End Of Transmission</td></tr> ENQ</td> 05</td> 005</td> 5</td> Enquiry</td></tr> ACK</td> 06</td> 006</td> 6</td> Acknowledge</td></tr> BEL</td> 07</td> 007</td> 7</td> Bell / Alert</td> \a</code></td></tr> BS</td> 08</td> 010</td> 8</td> BAckspace</td></tr> HT</td> 09</td> 011</td> 9</td> HOrizontal Tab</td> \t</code></td></tr> LF</td> 0a</td> 012</td> 10</td> Line Feed</td> \n</code></td></tr> VT</td> 0b</td> 013</td> 11</td> Vertical Tab</td> \v</code></td></tr> FF</td> 0c</td> 014</td> 12</td> Form Feed</td> \f</code></td></tr> CR</td> 0d</td> 015</td> 13</td> Carriage Return</td> \r</code></td></tr> SO</td> 0e</td> 016</td> 14</td> Shift Out</td></tr> SI</td> 0f</td> 017</td> 15</td> Shift In</td></tr> DLE</td> 10</td> 020</td> 16</td> Data Link Escape</td></tr> DC1</td> 11</td> 021</td> 17</td> Device Control 1</td></tr> DC2</td> 12</td> 022</td> 18</td> Device Control 2</td></tr> DC3</td> 13</td> 023</td> 19</td> Device Control 3</td></tr> DC4</td> 14</td> 024</td> 20</td> Device Control 4</td></tr> NAK</td> 15</td> 025</td> 21</td> Negative Acknowledge</td></tr> SYN</td> 16</td> 026</td> 22</td> Synchronous Idle</td></tr> ETB</td> 17</td> 027</td> 23</td> End Of Transmission Block</td></tr> CAN</td> 18</td> 030</td> 24</td> Cancel</td></tr> EM</td> 19</td> 031</td> 25</td> End Of Medium</td></tr> SUB</td> 1a</td> 032</td> 26</td> Substitute</td></tr> ESC</td> 1b</td> 033</td> 27</td> Escape</td> \e</code></td></tr> FS</td> 1c</td> 034</td> 28</td> File Seperator</td></tr> GS</td> 1d</td> 035</td> 29</td> Group Seperator</td></tr> RS</td> 1e</td> 036</td> 30</td> Record Seperator</td></tr> US</td> 1f</td> 037</td> 31</td> Unit Seperator</td></tr> </table> Note: The Names have been titlecased to make them easier to read. Notation</h2> Escaping in Scripting and Programming</h3> Sometimes in languages one can't use all the characters available in the source file, especially when using control characters. Encoding these as other characters is called escaping. This usually happens with text between double quotes "</code>. There are differences between languages and implementations but the general rules are: Special characters preceeded by a backslash \</code> are taken literally. (i.e. \"</code> or \\</code>) (The exact behaviour varies a lot by language.)</li> Latin characters preceeded by a backslash to generate control characters (i.e. \n</code>) (The C Esc column)</li> Hexadecimal value of the character prefixed by \x</code> (i.e. 0x1b</code>) (should work almost everywhere)</li> Octal encoding prefixed by just a backslash (i.e. \033</code>) (Use this when using printf</code> on the command line</a>)</li> </ul> Note: C Esc</code> here is short C-Escape as the C language apparantly started this way of excaping characters in strings. It is now used by almost all modern languages and in other contexts. i.e. printf</code>, sed</code>, etc. exact support may vary. Escaping in XML and HTML</h3> XML and html have their own way of escaping non-printable characters, this involves a sequence sandwiched between an ampersand &</code> and a semicolon ;</code>. In general Characters can be escaped using &#<dec>;</code> where <dec></code> is replaced by the decimal value associted with the character. (i.e, &</code> to encode an ampersand &</code>) There are also named escapes to make remembering them easier: Character</th> XML-escape</th></tr> &</code></td> &</code></td></tr> <</code></td> <</code></td></tr> ></code></td> ></code></td></tr> "</code></td> "</code></td></tr> </table> Control and Shift</h3> With the ASCII table control and shift keys an be implemented using simple addition and substraction. The lowercase character can be obtained by adding 32 (0x20) to the code of the corresponding uppercase character and the control key goes 64 (0x40) in the opposite direction. With that ctrl+c</kbd> maps to code 3 "End of Text". And ctrl+d</kbd> maps to "End of Transmission". You may know those shortcuts from the terminal, they hopefully make a bit more sense now. Control characters are sometimes written down/printed as the character one gets when adding 64 to the control characters value prefixed by an ^</code>. This maps escape to ^]</code> and the nullbyte to ^@</code>. (You have probably seen those when opening a binary file in a text editor.) Unicode</h2> Creating a unicode table andkeeping it updated is out of scope here, besides that: Wikipedia has a List of Unicode Characters</a> and there are the offical Unicode Character Code Charts</a>. (Plus a whole lot of other unicode tables out there.) Notation Hint: Unicode codepoints (that is the number assigned in the unicode standard, not the binary character representation) are written U+<hex></code> where <hex></code> is the codepint number in hexadecimal. Example: U+1f603</code> for 😃. There is also a little commandline tool called uni</code></a>, that is pretty good at providing a searchable unicode table. Shell Snippets 2022-10-16T00:00:00+00:00 The snippets on this site are intended for getting copied and being useful so please do that. Argument Parser</h2> A simple argument parser for bash. #!/bin/bash # Make the script fail when a command fails set -e # Settings variables go here FOO="foo" show_help() { cat <<EOF Usage: foo <<options>> <bar> Blubber blah blah ... EOF } while [ "$#" -gt 0 ]; do case "$1" in --foo) FOO="$2" shift 2;; --help) show_help; exit 0;; ) printf "Unknown option: %s\n" "$1"; exit 1;; esac done # Logic goes here echo "$FOO"</code></pre>set -e</code> and shift</code></h3> The set -e</code> option tells the shell to stop executing a script once a command fails. This is important when using a shift</code> in this kind of while loop where the loop exiting depends on the shift</code> working. Example: The shift may fail if it is told to move 2 arguments out (shift 2</code>) but there is only one given. In the example above that would be fulfilled when passing --foo</code> as the last argument (a relatively easy to make mistake). If the shift silently failed the argument counter would never reach 0 and the script would be stuck in an endless loop. Hook Helper</h2> This little function calls its argument as an ad-hoc shellscript if it is non-empty, if it returns a nonzero, one should call the default action. # on fail do default action run_hook() { [ -n "$1" ] && bash -c "$1" }</code></pre> Example of how this little helper is intended to be used</figcaption> export VARIABLE_WHICH_IS_USEFUL_FOR_HOOK="something" run_hook "$HOOK_COMMAND" || default_action</code></pre> </figure> An example of this being used</a> can be found in my dotfiles with the filedialog. (Fun fact: It was created for this script) Grep Regex Match Helper</h2> This function pipes its first argument into a grep -q</code> command which is given the rest of the arguments. matches() { TEXT="$1" shift 1 printf "%s\n" "$TEXT" | grep -q "$@" }</code></pre> Example usage: matches "$FOO" "^[0-9]$" && echo "FOO is a number! :)" || FOO=0</code> It can also be useful to implement some kind of heuristics based on which patterns occur in a text. Bash Arrays</h2> Bash has arrays so one doesn't have to care about delimiters too much. Note on compatibility: These arrays only work in bash which isn't isntalled everywhere (dash is a popular alternative). Please use #!/bin/bash</code> instead of #!/bin/sh</code></s> to indicate that your script requires bash to work. Create an array: empty_array=() array=('text' "can go" "$( echo here )" ";")</code></pre> Appending to an exisiting Array: array+=("foo") array+=("add" "multiple" "elements")</code></pre> These can then be accessed with 0 based indices: echo "${array[1]}" # prints out: can go</code></pre> To get the number of elements in an array: echo "${#array[*]}" echo "${#array[@]}"</code></pre> An array can expand to multiple arguments: printf "-> %s" "${array[@]}" # Compare with the output of: printf "-> %s" foo bar baz</code></pre> Which is also useful for for</code>-loops: for i in "${array[@]}" ; do echo "-> $i" done</code></pre> And to clean up: # remove element at index 1 uset array[1] # remove the whole array uset array</code></pre> There is more to bash arrays (including associative arrays) … Check out The Bash Reference Manual on Arrays </a> See Also</h2> ANSI Escape Sequences</a></li> Portable Shellscripting</a></li> </ul>

slatecave.net - Slatians Notebook

Caddy JSON log format

The SQLite strict Guide

Interesting Resources for building Search Engines

How Search works</h2> Articles that dive into how a Search Engine works under the hood.</p>

Acronyms on the Social Internet

Templating Zoo

Standards Zoo

Portable Shell-Scripting

What makes your shellscripts break?</h2> Before outlining how shellscripts can be made portable … why do they break?</p> This is a "you should consider", not a "you have to":</b> Depending on the purpose and intended audience of your script not all of the following may apply.</p>

POSIX Shell</h2> POSIX is a great thing, it standardises (amongst other things) shell syntax and command line utilities.</p> If you only use what POSIX defines in a way it is defined by POSIX your shellscripts will work across a wider range of systems without you having to do any extra work.</p>

ATmega328P and the Arduino UNO R3

XML XPath

Example Document</h2> For the Examples on this page the following XML-Document is assumed.</p>

Getting Attribute Text</h2> To get the text from an attribute one can prefix the attribute name with an @</code>.</p> Example:</b> To get all URLs from the link elements from the example document.</p>

Combine multiple XPaths</h2> Sometimes the result of one XPath isn't enough and you want the combined results of multiple XPaths. (Like a SQL union)</p> To achieve this you join multiple XPaths using a pipe |</code> character like this: /xpath1 | /xpath1</code></p>

Getting your Smartcards and Security Tokens working on Linux

SQL Cheatsheet

Git Cheatsheet(s)

Other Peoples Git Guides</h2> These are guides made by other people that I think are useful.</p>

Environment Variable Zoo

Reading and Writing Environment Variables</h2> These commands here should work in almost all shells inspired by the original sh, if you use something else you probably know why and how to use it.</p>

Graphical Environment</h2> DISPLAY</code> </dt> The id of your X-Display, on your laptop it is probably :0</code>. </dd> WAYLAND_DISPLAY</code> </dt> The name of your wayland socket file in $XDG_RUNTIME_DIR/</code>. </dd> XCURSOR_SIZE</code>, XCURSOR_THEME</code>, XCURSOR_PATH</code> </dt>

Mozilla</h2> These are used for Firefox, Thunderbird and other programs that are built on top of Firefox.</p>

Java</h2> _JAVA_AWT_WM_NONREPARENTING</code> </dt>

Kubernetes</h2> KUBECONFIG</code> </dt>

Fontconfig

What is Fontconfig?</h2> Fontconfig is a library for configuring and customising font access.</p> </blockquote> It is built into almost every program that runs on Linux and has a GUI (even in into old java software!) and makes it very easy to make configure fonts and then forget about it.</p>

ANSI Escape Sequences

Shell Null Termination / Separation

Speech Recognition Frontends

Overview</h2> The projects listed here fall into 3 broader categories.</p> Build your own voice something:</p>

What makes your shellscripts break?</h2>
Before outlining how shellscripts can be made portable … why do they break?</p>
This is a "you should consider", not a "you have to":</b> Depending on the purpose and intended audience of your script not all of the following may apply.</p>

Combine multiple XPaths</h2>
Sometimes the result of one XPath isn't enough and you want the combined results of multiple XPaths. (Like a SQL union)</p>
To achieve this you join multiple XPaths using a pipe `|</code> character like this: /xpath1 | /xpath1</code></p>`