finish up rss feeds

3 files changed, 562 insertions, 1 deletions

diff --git a/docs/index.xml b/docs/index.xml
new file mode 100644
index 0000000..cb23ad8
--- /dev/null
+++ b/docs/index.xml
@@ -0,0 +1,519 @@
+<rss version="2.0">
+    <channel>
+      <title>nerdypepper's μblog</title>
+      <link>https://peppe.rs</link>
+      <description>programming, design, software</description>
+      <link href="https://peppe.rs/index.xml" rel="self" type="application/xml"/>
+      <image>
+        <title>nerdypepper logo</title>
+        <url>https://u.peppe.rs/n.png</url>
+        <link>https://peppe.rs</link>
+      </image>
+      <language>en-us</language>
+      <copyright>Creative Commons BY-NC-SA 4.0</copyright>
+      <item>
+<title>Termux Tandem</title>
+<description><p>I learnt about <code>termux</code> from a friend on IRC recently.
+It looked super gimmicky to me at first, but it eventually
+proved to be useful. Here&#39;s what I use it for:</p>
+
+<h3 id="rsync">rsync</h3>
+
+<p>Ever since I degoogled my android device, syncing files
+between my phone and my PC has always been a pain. I&#8217;m
+looking at you MTP. But, with <code>termux</code> and <code>sshd</code> all set up,
+it&#39;s as simple as:</p>
+
+<pre><code>$ arp
+Address         HWtype  HWad ...
+192.168.43.187  ether   d0:0 ...
+
+$ rsync -avz 192.168.43.187:~&#47;frogs ~&#47;pics&#47;frogs
+</code></pre>
+
+<h3 id="ssh%20&amp;#38;%20tmux">ssh &#38; tmux</h3>
+
+<p>My phone doubles as a secondary view into my main machine
+with <code>ssh</code> and <code>tmux</code>. When I am away from my PC (read:
+sitting across the room), I check build status and IRC
+messages by <code>ssh</code>ing into a tmux session running the said
+build or weechat.</p>
+
+<h3 id="file%20uploads">file uploads</h3>
+
+<p>Not being able to access my (ssh-only) file host was
+crippling. With a <code>bash</code> instance on my phone, I just copied
+over my ssh keys, and popped in a file upload script (a
+glorified <code>scp</code>). Now I just have to figure out a way to
+clean up these file names &#8230;</p>
+
+<pre><code>~&#47;storage&#47;pictures&#47; $ ls
+02muf5g7b2i41.jpg  7alt3cwg77841.jpg  cl4bsrge7id11.png
+mtZabXG.jpg        p8d5c584f2841.jpg  vjUxGjq.jpg
+</code></pre>
+
+<h3 id="cmus">cmus</h3>
+
+<p>Alright, I don&#39;t really listen to music via <code>cmus</code>, but I
+did use it a couple times when my default music player was
+acting up. <code>cmus</code> is a viable option:</p>
+
+<p><a href="https://u.peppe.rs/CP.jpg"><img src="https://u.peppe.rs/CP.jpg" alt="cmus.png" /></a></p></description>
+<link>https://peppe.rs/posts/termux_tandem/</link>
+<pubDate>Sun, 08 Mar 20 16:47:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/termux_tandem/</guid>
+</item>
+<item>
+<title>Call To ARMs</title>
+<description><p>My 4th semester involves ARM programming. And proprietary
+tooling (Keil C). But we don&#39;t do that here.</p>
+
+<h3 id="Building">Building</h3>
+
+<p>Assembling and linking ARM binaries on non-ARM architecture
+devices is fairly trivial. I went along with the GNU cross
+bare metal toolchain binutils, which provides <code>arm-as</code> and
+<code>arm-ld</code> (among a bunch of other utils that I don&#39;t care
+about for now). </p>
+
+<p>Assemble <code>.s</code> files with:</p>
+
+<pre><code class="language-shell">arm-none-eabi-as main.s -g -march=armv8.1-a -o main.out
+</code></pre>
+
+<p>The <code>-g</code> flag generates extra debugging information that
+<code>gdb</code> picks up. The <code>-march</code> option establishes target
+architecture.</p>
+
+<p>Link <code>.o</code> files with:</p>
+
+<pre><code class="language-shell">arm-none-eabi-ld main.out -o main
+</code></pre>
+
+<h3 id="Running%20(and%20Debugging)">Running (and Debugging)</h3>
+
+<p>Things get interesting here. <code>gdb</code> on your x86 machine
+cannot read nor execute binaries compiled for ARM. So, we
+simulate an ARM processor using <code>qemu</code>. Now qemu allows you
+to run <code>gdbserver</code> on startup. Connecting our local <code>gdb</code>
+instance to <code>gdbserver</code> gives us a view into the program&#8217;s
+execution. Easy!</p>
+
+<p>Run <code>qemu</code>, with <code>gdbserver</code> on port <code>1234</code>, with our ARM
+binary, <code>main</code>:</p>
+
+<pre><code class="language-shell">qemu-arm -singlestep -g 1234 main
+</code></pre>
+
+<p>Start up <code>gdb</code> on your machine, and connect to <code>qemu</code>&#8217;s
+<code>gdbserver</code>:</p>
+
+<pre><code>(gdb) set architecture armv8-a
+(gdb) target remote localhost:1234
+(gdb) file main
+Reading symbols from main...  # yay!
+</code></pre>
+
+<h3 id="GDB%20Enhanced">GDB Enhanced</h3>
+
+<p><code>gdb</code> is cool, but it&#39;s not nearly as comfortable as well
+fleshed out emulators&#47;IDEs like Keil. Watching registers,
+CPSR and memory chunks update <em>is</em> pretty fun. </p>
+
+<p>I came across <code>gdb</code>&#39;s TUI mode (hit <code>C-x C-a</code> or type <code>tui
+enable</code> at the prompt). TUI mode is a godsend. It highlights
+the current line of execution, shows you disassembly
+outputs, updated registers, active breakpoints and more.</p>
+
+<p><em>But</em>, it is an absolute eyesore.</p>
+
+<p>Say hello to <a href="https://github.com/hugsy/gef">GEF</a>! &#8220;GDB
+Enhanced Features&#8221; teaches our old dog some cool new tricks.
+Here are some additions that made my ARM debugging
+experience loads better:</p>
+
+<ul>
+<li>Memory watches</li>
+<li>Register watches, with up to 7 levels of deref (overkill,
+I agree)</li>
+<li>Stack tracing</li>
+</ul>
+
+<p>And it&#39;s pretty! See for yourself:</p>
+
+<p><a href="https://u.peppe.rs/wq.png"><img src="https://u.peppe.rs/wq.png" alt="gef.png" /></a></p>
+
+<h3 id="Editing">Editing</h3>
+
+<p>Vim, with <code>syntax off</code> because it
+dosen&#39;t handle GNU ARM syntax too well.</p></description>
+<link>https://peppe.rs/posts/call_to_ARMs/</link>
+<pubDate>Fri, 07 Feb 20 18:30:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/call_to_ARMs/</guid>
+</item>
+<item>
+<title>Color Conundrum</title>
+<description><p>This piece aims to highlight (pun intended) some of the
+reasons behind my <a href="https://u.peppe.rs/bF.png">color
+free</a> editor setup.</p>
+
+<p>Imagine highlighting an entire book because <em>all</em> of it is
+important. That is exactly what (most) syntax highlighting
+does. It is difficult for the human eye to filter out noise
+in rainbow barf. Use color to draw attention, not diverge
+it.</p>
+
+<p>At the same time, a book devoid of color is <em>boring!</em> What
+is the takeaway from this 10 line paragraph? What are the
+technical terms used?</p>
+
+<p>Prose and code are certainly different, but the fickle
+minded human eye is the same. The eye constantly looks for a
+frame of reference, a focal point. It grows tired when it
+can&#39;t find one.</p>
+
+<p>The following comparison does a better job of explaining
+(none, ample and over-the-top highlighting, from left to
+right):</p>
+
+<p><a href="https://u.peppe.rs/lt.png"><img src="https://u.peppe.rs/lt.png" alt="hi.png" /></a></p>
+
+<p>Without highlighting (far left), it is hard to differentiate
+between comments and code! The florid color scheme (far
+right) is no good either, it contains too many attention
+grabbers. The center sample is a healthy balance of both.
+Function calls and constants stand out, and repetitive
+keywords and other noise (<code>let</code>, <code>as</code>) are mildly dimmed
+out. Comments and non-code text (sign column, status text)
+are dimmed further.</p>
+
+<p>I&#39;ll stop myself before I rant about color contrast and
+combinations.</p></description>
+<link>https://peppe.rs/posts/color_conundrum/</link>
+<pubDate>Mon, 30 Dec 19 18:30:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/color_conundrum/</guid>
+</item>
+<item>
+<title>Static Sites With Bash</title>
+<description><p>After going through a bunch of static site generators
+(<a href="https://blog.getpelican.com/">pelican</a>,
+<a href="https://gohugo.io">hugo</a>,
+<a href="https://github.com/icyphox/vite">vite</a>), I decided to roll
+my own. If you are more of the &#8216;show me the code&#8217; kinda guy,
+<a href="https://github.com/nerdypepper/site">here</a> you go.</p>
+
+<h3 id="Text%20formatting">Text formatting</h3>
+
+<p>I chose to write in markdown, and convert
+to html with <a href="https://kristaps.bsd.lv/lowdown/">lowdown</a>.</p>
+
+<h3 id="Directory%20structure">Directory structure</h3>
+
+<p>I host my site on GitHub pages, so
+<code>docs&#47;</code> has to be the entry point. Markdown formatted posts
+go into <code>posts&#47;</code>, get converted into html, and end up in
+<code>docs&#47;index.html</code>, something like this:</p>
+
+<pre><code>posts=$(ls -t .&#47;posts)     # chronological order!
+for f in $posts; do
+    file=&#34;.&#47;posts&#47;&#34;$f      # `ls` mangled our file paths
+    echo &#34;generating post $file&#34;
+
+    html=$(lowdown &#34;$file&#34;)
+    echo -e &#34;html&#34; &#62;&#62; docs&#47;index.html
+done
+</code></pre>
+
+<h3 id="Assets">Assets</h3>
+
+<p>Most static site generators recommend dropping image
+assets into the site source itself. That does have it&#8217;s
+merits, but I prefer hosting images separately:</p>
+
+<pre><code># strip file extension
+ext=&#34;${1##*.}&#34;
+
+# generate a random file name
+id=$( cat &#47;dev&#47;urandom | tr -dc &#39;a-zA-Z0-9&#39; | fold -w 2 | head -n 1 )
+id=&#34;$id.$ext&#34;
+
+# copy to my file host
+scp -P 443 &#34;$1&#34; emerald:files&#47;&#34;$id&#34; 
+echo &#34;https:&#47;&#47;u.peppe.rs&#47;$id&#34;
+</code></pre>
+
+<h3 id="Templating">Templating</h3>
+
+<p><a href="https://github.com/NerdyPepper/site/blob/master/generate.sh"><code>generate.sh</code></a>
+brings the above bits and pieces together (with some extra
+cruft to avoid javascript).  It uses <code>sed</code> to produce nice
+titles from the file names (removes underscores,
+title-case), and <code>date(1)</code> to add the date to each post
+listing!</p></description>
+<link>https://peppe.rs/posts/static_sites_with_bash/</link>
+<pubDate>Fri, 22 Nov 19 18:30:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/static_sites_with_bash/</guid>
+</item>
+<item>
+<title>My Setup</title>
+<description><p>Decided to do one of these because everyone does one of
+these.</p>
+
+<p><img src="https://u.peppe.rs/Hb.png" alt="scrot" /></p>
+
+<p>My entire setup is managed with GNU <code>stow</code>, making it easier
+to replicate on fresh installations. You can find my
+configuration files on <a href="https://github.com/nerdypepper">GitHub</a>.</p>
+
+<p>I run Void Linux (glibc) on my
+<a href="https://store.hp.com/us/en/mdp/laptops/envy-13">HP Envy 13&#8221; (2018)</a>.
+To keep things simple, I run a raw X session with <code>2bwm</code> as my
+window manager, along with <code>dunst</code> (notification daemon) and
+Sam&#39;s <a href="https://github.com/sdhand/compton"><code>compton</code></a>
+(compositor) fork.</p>
+
+<p>I am a fan of GNU tools, so I use <code>bash</code> as my shell, and
+<code>coreutils</code> to manage files, archives, strings, paths etc. I
+edit files with <code>vim</code>, chat with <code>weechat</code>, listen to music
+with <code>cmus</code>, monitor processes with <code>htop</code>, manage sessions
+with <code>tmux</code>, read pdfs in <code>zathura</code>. I rarely ever leave
+the comfort of my terminal emulator, <code>urxvt</code>.</p>
+
+<p>Most of my academic typesetting is done with TeX, and
+compiled with <code>xelatex</code>. Other <em>fun</em> documents are made with
+GIMP :).</p></description>
+<link>https://peppe.rs/posts/my_setup/</link>
+<pubDate>Wed, 06 Nov 19 18:30:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/my_setup/</guid>
+</item>
+<item>
+<title>WPA Woes</title>
+<description><p>I finally got around to installing Void GNU&#47;Linux on my main
+computer. Rolling release, non-systemd, need I say more?</p>
+
+<p>As with all GNU&#47;Linux distributions, wireless networks had
+me in a fix. If you can see this post, it means I&#39;ve managed
+to get online. It turns out, <code>wpa_supplicant</code> was detecting the
+wrong interface by default (does it ever select the right
+one?). Let us fix that:</p>
+
+<pre><code>$ sudo rm -r &#47;var&#47;service&#47;wpa_supplicant
+$ sudo killall dhcpcd
+</code></pre>
+
+<p>What is the right interface though?</p>
+
+<pre><code>$ iw dev
+   ...
+   Interface wlp2s0
+   ...
+</code></pre>
+
+<p>Aha! Let us run <code>wpa_supplicant</code> on that interface, as a
+background process:</p>
+
+<pre><code>$ sudo wpa_supplicant -B -i wlp2s0 -c &#47;etc&#47;wpa_supplicant&#47;wpa_supplicant.conf
+$ sudo dhcpcd -B wlp2s0
+$ ping google.com
+PING ...
+</code></pre>
+
+<p>Yay! Make those changes perpetual by enabling the service:</p>
+
+<pre><code>------------------------------------------------------
+# Add these to &#47;etc&#47;wpa_supplicant&#47;wpa_supplicant.conf
+OPTS=&#34;-B&#34;
+WPA_INTERFACE=&#34;wlp2s0&#34;
+------------------------------------------------------
+$ sudo ln -s &#47;etc&#47;sv&#47;wpa_supplicant &#47;var&#47;service&#47;
+$ sudo ln -s &#47;etc&#47;sv&#47;dhcpcd &#47;var&#47;service&#47;
+$ sudo sv restart wpa_supplicant
+$ sudo sv restart dhcpcd
+</code></pre></description>
+<link>https://peppe.rs/posts/WPA_woes/</link>
+<pubDate>Sat, 12 Oct 19 16:18:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/WPA_woes/</guid>
+</item>
+<item>
+<title>Bye Bye BDFs</title>
+<description><p>Glyph Bitmap Distribution Format is no more, as the creators of
+<a href="https://pango.org">Pango</a>, one of the most widely used text rendering
+libraries,
+<a href="https://blogs.gnome.org/mclasen/2019/05/25/pango-future-directions/">announced</a>
+their plans for Pango 1.44.</p>
+
+<p>Until recently, Pango used FreeType to draw fonts. They will be moving over
+to <a href="https://harfbuzz.org">Harfbuzz</a>, an evolution of FreeType.</p>
+
+<p><em>Why?</em></p>
+
+<p>In short, FreeType was hard to work with. It required complex logic, and 
+provided no advantage over Harfbuzz (other than being able to fetch
+opentype metrics with ease).</p>
+
+<p>Upgrading to Pango v1.44 will break your GTK applications (if you use a
+<code>bdf</code>&#47;<code>pcf</code> bitmap font). Harfbuzz <em>does</em> support bitmap-only OpenType fonts,
+<code>otb</code>s. Convert your existing fonts over to <code>otb</code>s using
+<a href="https://fontforge.github.io">FontForge</a>. It is to be noted that applications
+such as <code>xterm</code> and <code>rxvt</code> use <code>xft</code> (X FreeType) to render fonts, and will
+remain unaffected by the update.</p>
+
+<p>Both <a href="https://github.com/nerdypepper/scientifica">scientifica</a> and
+<a href="https://github.com/nerdypepper/curie">curie</a> will soon ship with bitmap-only
+OpenType font formats.</p></description>
+<link>https://peppe.rs/posts/bye_bye_BDFs/</link>
+<pubDate>Wed, 07 Aug 19 12:31:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/bye_bye_BDFs/</guid>
+</item>
+<item>
+<title>Onivim Sucks</title>
+<description><p><a href="https://v2.onivim.io">Onivim</a> is a &#8216;modern modal editor&#8217;, combining fancy
+interface and language features with vim-style modal editing. What&#39;s wrong you
+ask?</p>
+
+<p>Apart from <a href="https://github.com/onivim/oni2/issues/550">buggy syntax highlighting</a>, 
+<a href="https://github.com/onivim/oni2/issues/519">broken scrolling</a> and
+<a href="https://github.com/onivim/oni2/issues?q=is%3Aissue+label%3A%22daily+editor+blocker%22+is%3Aopen">others</a>,
+Onivim is <strong>proprietary</strong> software. It is licensed under a commercial 
+<a href="https://github.com/onivim/oni1/blob/master/Outrun-Labs-EULA-v1.1.md">end user agreement license</a>,
+which prohibits redistribution in both object code and source code formats.</p>
+
+<p>Onivim&#39;s core editor logic (bits that belong to vim), have been separated from
+the interface, into <a href="https://github.com/onivim/libvim">libvim</a>. libvim is
+licensed under MIT, which means, this &#8216;extension&#8217; of vim is perfectly in
+adherence to <a href="http://vimdoc.sourceforge.net/htmldoc/uganda.html#license">vim&#39;s license text</a>! 
+Outrun Labs are exploiting this loophole (distributing vim as a library) to
+commercialize Onivim.</p>
+
+<p>Onivim&#39;s source code is available on <a href="https://github.com/onivim/oni2">GitHub</a>.
+They do mention that the source code trickles down to the
+<a href="https://github.com/onivim/oni2-mit">oni2-mit</a> repository, which (not yet) contains
+MIT-licensed code, <strong>18 months</strong> after each commit to the original repository.</p>
+
+<p>Want to contribute to Onivim? Don&#39;t. They make a profit out of your contributions.
+Currently, Onivim is priced at $19.99, &#8216;pre-alpha&#8217; pricing which is 80% off the
+final price! If you are on the lookout for an editor, I would suggest using
+<a href="https://vim.org">Vim</a>, charity ware that actually works, and costs $100 lesser.</p></description>
+<link>https://peppe.rs/posts/onivim_sucks/</link>
+<pubDate>Fri, 02 Aug 19 12:31:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/onivim_sucks/</guid>
+</item>
+<item>
+<title>Bash Harder With Vim</title>
+<description><p>Bash is tricky, don&#39;t let your editor get in your way. Here&#39;s a couple of neat
+additions you could make to your <code>vimrc</code> for a better shell programming
+experience.</p>
+
+<h3 id="Man%20pages%20inside%20vim">Man pages inside vim</h3>
+
+<p>Source this script to get started:  </p>
+
+<pre><code>runtime ftplugin&#47;man.vim
+</code></pre>
+
+<p>Now, you can open manpages inside vim with <code>:Man</code>! It adds nicer syntax highlighting
+and the ability to jump around with <code>Ctrl-]</code> and <code>Ctrl-T</code>.</p>
+
+<p>By default, the manpage is opened in a horizontal split, I prefer using a new tab:</p>
+
+<pre><code>let g:ft_man_open_mode = &#39;tab&#39;
+</code></pre>
+
+<h3 id="Scratchpad%20to%20test%20your%20commands">Scratchpad to test your commands</h3>
+
+<p>I often test my <code>sed</code> substitutions, here is
+a sample from the script used to generate this site:  </p>
+
+<pre><code># a substitution to convert snake_case to Title Case With Spaces
+echo &#34;$1&#34; | sed -E -e &#34;s&#47;\..+$&#47;&#47;g&#34;  -e &#34;s&#47;_(.)&#47; \u\1&#47;g&#34; -e &#34;s&#47;^(.)&#47;\u\1&#47;g&#34;
+</code></pre>
+
+<p>Instead of dropping into a new shell, just test it out directly from vim!</p>
+
+<ul>
+<li><p>Yank the line into a register:</p>
+
+<pre><code>yy
+</code></pre></li>
+<li><p>Paste it into the command-line window:</p>
+
+<pre><code>q:p
+</code></pre></li>
+<li><p>Make edits as required:</p>
+
+<pre><code>syntax off            # previously run commands
+edit index.html       # in a buffer!
+w | so %
+!echo &#34;new_post.md&#34; | sed -E -e &#34;s&#47;\..+$&#47;&#47;g&#34;  --snip--
+^--- note the use of &#39;!&#39;
+</code></pre></li>
+<li><p>Hit enter with the cursor on the line containing your command!</p>
+
+<pre><code>$ vim
+New Post         # output
+Press ENTER or type command to continue
+</code></pre></li>
+</ul></description>
+<link>https://peppe.rs/posts/bash_harder_with_vim/</link>
+<pubDate>Wed, 31 Jul 19 06:30:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/bash_harder_with_vim/</guid>
+</item>
+<item>
+<title>Hold Position!</title>
+<description><p>Often times, when I run a vim command that makes &#8220;big&#8221; changes to a file (a
+macro or a <code>:vimgrep</code> command) I lose my original position and feel disoriented.</p>
+
+<p><em>Save position with <code>winsaveview()</code>!</em></p>
+
+<p>The <code>winsaveview()</code> command returns a <code>Dictionary</code> that contains information
+about the view of the current window. This includes the cursor line number,
+cursor coloumn, the top most line in the window and a couple of other values,
+none of which concern us.</p>
+
+<p>Before running our command (one that jumps around the buffer, a lot), we save
+our view, and restore it once its done, with <code>winrestview</code>.</p>
+
+<pre><code>let view = winsaveview()
+s&#47;\s\+$&#47;&#47;gc              &#34; find and (confirm) replace trailing blanks
+winrestview(view)        &#34; restore our original view!
+</code></pre>
+
+<p>It might seem a little overkill in the above example, just use `` (double
+backticks) instead, but it comes in handy when you run your file through
+heavier filtering.</p></description>
+<link>https://peppe.rs/posts/hold_position!/</link>
+<pubDate>Tue, 30 Jul 19 12:31:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/hold_position!/</guid>
+</item>
+<item>
+<title>Get Better At Yanking And Putting In Vim</title>
+<description><ol start="1">
+<li><p>reselecting previously selected text (i use this to fix botched selections):</p>
+
+<pre><code>gv  &#34; :h gv for more
+    &#34; you can use `o` in visual mode to go to the `Other` end of the selection
+    &#34; use a motion to fix the selection
+</code></pre></li>
+<li><p>reselecting previously yanked text:</p>
+
+<pre><code>`[v`]
+`[         &#34; marks the beginning of the previously yanked text   :h `[
+`]         &#34; marks the end                                       :h `]
+ v         &#34; visual select everything in between
+
+nnoremap gb `[v`]    &#34; &#34;a quick map to perform the above
+</code></pre></li>
+<li><p>pasting and indenting text (in one go):</p>
+
+<pre><code>]p   &#34; put (p) and adjust indent to current line
+]P   &#34; put the text before the cursor (P) and adjust indent to current line
+</code></pre></li>
+</ol></description>
+<link>https://peppe.rs/posts/get_better_at_yanking_and_putting_in_vim/</link>
+<pubDate>Mon, 29 Jul 19 12:31:00 +0000</pubDate>
+<guid>https://peppe.rs/posts/get_better_at_yanking_and_putting_in_vim/</guid>
+</item>
+
+
+    </channel>
+</rss>
diff --git a/generate.sh b/generate.sh
index ccae2dd..a836eb1 100755
--- a/generate.sh
+++ b/generate.sh
@@ -116,9 +116,14 @@ for f in $posts; do
         title="$post_title" \
         read_time="$r_time" \
         height="$height"
-
 done
 
+# generate rss feeds
+echo "generating RSS feeds ..."
+esh -s /bin/bash \
+    -o "./docs/index.xml" \
+    "rss.esh"
+
 cat >> ./docs/index.html << EOF
     </table>
     <div class="separator"></div>
diff --git a/rss.esh b/rss.esh
new file mode 100644
index 0000000..941ccd7
--- /dev/null
+++ b/rss.esh
@@ -0,0 +1,37 @@
+<rss version="2.0">
+    <channel>
+      <title>nerdypepper's μblog</title>
+      <link>https://peppe.rs</link>
+      <description>programming, design, software</description>
+      <link href="https://peppe.rs/index.xml" rel="self" type="application/xml"/>
+      <image>
+        <title>nerdypepper logo</title>
+        <url>https://u.peppe.rs/n.png</url>
+        <link>https://peppe.rs</link>
+      </image>
+      <language>en-us</language>
+      <copyright>Creative Commons BY-NC-SA 4.0</copyright>
+      <% for f in `ls -t ./posts`; do
+          file="./posts/"$f
+          post_date=$(date -u -r "$file" "+%a, %d %b %y %H:%M:00 %z")
+          html=$(lowdown "$file")
+          id="${file##*/}"
+          id="${id%.*}"
+          post_title=$(echo "$id" | sed -E -e "s/\..+$//g"  -e "s/_(.)/ \u\1/g" -e "s/^(.)/\u\1/g")
+          post_link="https://peppe.rs/posts/$id/"
+
+          echo "<item>"
+
+          echo "<title>$post_title</title>"
+          echo "<description>$html</description>"
+          echo "<link>$post_link</link>"
+          echo "<pubDate>$post_date</pubDate>"
+          echo "<guid>$post_link</guid>"
+
+          echo "</item>"
+
+          done
+      %>
+
+    </channel>
+</rss>