[NTG-context] ConTeXt source browser website update

[Taco Hoekwater]

Hi all,

I just replaced the software of https://source.contextgarden.net/ by a Lua implementation that uses Hans Hagen’s SciTE lexers for syntax highlighting and a Lua-generated search index.

There is no official documentation yet (because I may want to change some stuff still) but here is some relevant information:

* The search field uses a Lua string pattern to match a single word (no multi-word search as of yet). 

  See https://www.lua.org/manual/5.3/manual.html#6.4.1 for the syntax of lua string patterns.

* The search index only contains 7-bit ASCII ‘words’ that contain at least one of 'a-zA-Z’ (writing an LPEG for unicode matching is a tricky and probably would be quite slow even if I could make it work). 

  Still, this should be ok-ish because most of the ConTeXt source is ASCII, except for the interface files.

* The files context-en.xml, mtxrun{.lua}, and all of the files in the tex/context/patterns subtree are skipped in the search index. Also skipped are words shorter than 3 bytes, lua keywords, and words having more than (approx) 1500 hits

* On the plus side, because of the SciTE lexer, “contextversion’ and “\contextversion” are separate entries in the search index. And you can search for “\@@kldirection”.

* For text file types that have no SciTE lexer there is no syntax highlighting, and binary files are ignored altogether.

The rest of the interface should be self-explanatory.

Suggestions for improvements are welcome.


Index statistics:

4085 files processed, 178 skipped, 3744 with matches
120733 words found
119154 words exported to word.idx
37928 comment words found
36880 words exported to comment.idx
26756 documentation words found
26018 words exported to docs.idx

The word.idx is roughly 11 megabytes.

Have fun,

Taco