<div dir="ltr">On Mon, Nov 13, 2017 at 11:50 PM, Julian Andres Klode <span dir="ltr"><<a href="mailto:jak@debian.org" target="_blank">jak@debian.org</a>></span> wrote:<br><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">(forwarding this to ubuntu-devel-discuss and Zygmunt)<br> <span><br> On Mon, Nov 13, 2017 at 10:33:39PM -0800, Shawn Landden wrote:<br> > Package: command-not-found<br> > Severity: wishlist<br> ><br> > I re-wrote command-not-found to get rid of the python dependancy, and<br> > to reduce the database size, as to reduce memory usage.<br> ><br> > <a href="https://github.com/shawnl/command-not-found" rel="noreferrer" target="_blank">https://github.com/shawnl/comm<wbr>and-not-found</a><br> ><br> > I was preparing to upload it to mentors as command-not-found-ng<br> <br> </span>I also rewrote it years ago, but using the same database format,<br> just in C. It was a lot faster. I don't understand the memory usage<br> bit - it should not matter how large the database is, it's memory<br> mapped, and not read into memory, as such memory usage should be<br> roughly constant.<br> <br></blockquote><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> Questions/Comments for your approach:<br> <br> * Did you test your format on a slow HDD with caches dropped? It<br> must not be slower than the Python one (that one is way too slow<br> already) - I did, it seems to be faster (0.4 vs 0.68 seconds)<br> - I believe the database-based C rewrite was even much faster,<br> though.<br></blockquote><div>Yes, as the disk IO is all the time, I think its best to keep the file size small. Then it has more chance of staying in memory. <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * update-command-not-found should use apt-get indextargets<br></blockquote><div>fixed <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * You don't store components, hence you cannot tell people to enable<br> component. That's a very important use case for Ubuntu, where<br> not all components are enabled by default, but the database is<br> shipped in the package.<br> <br> You could just append /<component> to each package name I think,<br> and strip it away when displaying.<br></blockquote><div>fixed <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * You should use getopt_long() to parse command-line options, and<br> support -h, --help :)<br></blockquote><div>fixed <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * pts_lbsearch belongs into usr/lib/..., not usr/share/...<br></blockquote><div>the seperate binary is gone <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <br> * You don't implement a closest matches function:<br> <br> $ command-not-found thunderbrd<br> No command 'thunderbrd' found, did you mean:<br> Command 'thunderbird' from package 'thunderbird' (main)<br> thunderbrd: command not found<br> $ ./command-not-found thunderbrd<br> thunderbrd: command not found<br> <br> This one is really important. People do make typos or misremember<br> command names, so the tool needs to be able to deal with that<br> <br> Should be easy to implement though, although you might have to<br> search multiple times - once for each alternative. All you need is<br> <br> def similar_words(word):<br> """ return a set with spelling1 distance alternative spellings<br> <br> based on <a href="http://norvig.com/spell-correct.html" rel="noreferrer" target="_blank">http://norvig.com/spell-correc<wbr>t.html</a>"""<br> alphabet = 'abcdefghijklmnopqrstuvwxyz-_'<br> s = [(word[:i], word[i:]) for i in range(len(word) + 1)]<br> deletes = [a + b[1:] for a, b in s if b]<br> transposes = [a + b[1] + b[0] + b[2:] for a, b in s if len(b)>1]<br> replaces = [a + c + b[1:] for a, b in s for c in alphabet if b]<br> inserts = [a + c + b for a, b in s for c in alphabet]<br> return set(deletes + transposes + replaces + inserts)<br> <br> And search for what that returns. And you don't need to search for those<br> at all if you have a direct match.<br> <br></blockquote><div>fixed, and I believe bit-for-bit identical <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * It needs to be translated - also very important.<br></blockquote><div>I made a pot file and used translations from the python version, but I can't get my app to look for translations (as examined through strace). I read the gettext manual and do not know what I am doing wrong. <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> <br> * You need to Conflict with command-not-found and not Break AFAIUI<br> <br></blockquote><div>fixed <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * You should not depend on grep, sed, coreutils, they are Essential.<br> <br></blockquote><div>fixed, now it uses ruby as my shell was hacky. <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * You do have to Depend on apt-file, as that configures apt to download<br> the Contents files<br> <br></blockquote><div>fixed <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> * You should not have identifiers starting with _ in the program, these<br> are reserved for the C implementation (like _cleanup_free_).<br> <br></blockquote><div>fixed <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> Yes, and these are basically the same reasons my C prototype is<br> not in the archive. Also, I did not put a lot of work into it, as<br> I was waiting for PackageKit to take that over, but that was not<br> done yet.<br> <br> I think it's a worthwhile approach, and I can see it replacing<br> command-not-found if those tiny issues have been fixed. Then you<br> could also avoid the -ng moniker, and just take over the main<br> package (if Zygmunt does not mind), which also avoids a month<br> long NEW process :)<br> <span class="m_2069971577309996656HOEnZb"><font color="#888888"><br> --<br> Debian Developer - <a href="http://deb.li/jak" rel="noreferrer" target="_blank">deb.li/jak</a> | <a href="http://jak-linux.org" rel="noreferrer" target="_blank">jak-linux.org</a> - free software dev<br> Ubuntu Core Developer de, en speaker<br> </font></span></blockquote></div><br></div></div>