A critical assessment of SILE's classes and packages in 2025

[Omikhleia]

In #2330, @jodros wrote:

... to make SILE simpler: get rid of all these useless packages and classes (there are some which are merely drafts and shouldn't be in the main branch), and then simply focus on improving technical matters ...

And I answered:

Surely, one could do a critical assessment of other classes and packages ...

So here are my personal assessments of the classes and packages currently in SILE's main distribution, as of November 2025, with some suggestions on what to do with them.

This is of course fairly subjective, and open to discussion. It also assumes we can have contributors willing to take on the work on some of the items marked for improvement...

Classes Assessment

Class	Notes	Personal opinion
bible.lua	Why?	🤔 See note (1) below
book.lua	Limited	💬 See related 2025 assessment (#2330)
diglot.lua	Why?	🤔 See note (1) below
docbook.lua	Broken (#1338)	💀 Abandon and remove (with updates to the manual too)
jbook.lua	Undocumented	📚 See note (2) below
jplain.lua	Undocumented	📚 See note (2) below
letter.lua	Undocumented (#1779) and naive	💀 Abandon and remove, or move to 3rd-party
pecha.lua	Undocumented	📚 See note (2) below
plain.lua	-	✏️ Move TeX-style commands to a package (e.g. see re·sil·ient's approach)
tbook.lua	Undocumented	📚 See note (2) below
tplain.lua	Undocumented	📚 See note (2) below
triglot.lua	Why?	📚 See note (1) below

Notes:

• (1) It's probably quite dubious to have such things as classes, non-derivative from a generic book/document class... It's also dubious to have such specialized classes in the main distribution, while a 3rd-party approach could be more appropriate...
• (2) There are several open issues... But at least some reverse-engineering and documentation is needed if these are to be kept alive...

Packages Assessment

Package	Notes	Personal opinion
autodoc (!!)	Dubious	🤔 Partial loading of package bits is NOT a sane approach IMHO...
background	Good but limited	✏️ Could benefit from more options for images (no redimension, positioning, panning, etc.) EDIT See #2346
balanced-frames	Related issues (#118) + barely working	💀 Abandon and/or reconsider?
bibtex (*)	Ongoing work in progress	✏️ Continue work but some contributors would be great...
bidi (!!)	Quite broken	✏️ See issues and PRs (#2268,#1796,#466,#83,#77)...
boustrophedon	Related issue (#2123)	✏️ Do BCP47 etc. and fix the issue
break-firstfit (!!)	Useless/Unused?	💀 Abandon and remove
chapterverse	Why? Too specialized	See note (1) above on bible class...
chordmode	Okayish	-
color	Okayish	✏️ What about more CSS-style support, ICC generalizatio (#444), conversion tooling...
color-fonts	Related issues (#501,#1261,#2232)	✏️ Consider fixes and improvements as per issues... Maybe a larger shaper refactor...
complex-spaces	Unclear	📚 Clarify...
converters	Limited, probably unused, but okayish...	-
counters	Okayish (but see #1248)	✏️ Too many direct access to internal/scratch stuff in the code base to make it robust...
cropmarks	Related issue (#432) but okayish	-
date	OS-specific but okayish...	-
dropcaps	Okayish	-
features	?	(I don't really know about this one)
folio (!)	Naive but okeyish	-
font-fallback	Unsatisfying (#1543)	✏️ Consider fixes and improvements as per issue...
footnotes (!)	Okayish	-
frametricks	?	(A mixed bad of things, not assessed here...)
grid (!!)	Crap (#1720,#826,#533)	🤔 Quite messy, how to proceed?
gutenberg (!!)	Broken (#724,#1227)	💀 Abandon and remove
hanmenkyoshi	?	(I don't have an opinion on this one)
ifattop	Broken (#342)	🤔 No idea, I never used it
image	Okayish	✏️ Could benefit from more options (max-width/height, PDF/UA support, etc.)
indexer	Okayish	-
infonode	Related issue (#1816)	✏️ Clarify and fix as per issue
inputfilter	Naive but okayish	-
insertions	Insertions logic is broken (#1828,#619,#420,#373,#214)	🤔 Quite messy, how to proceed?
leaders	Okayish	-
linespacing (!!)	Crap (#882,#823)	💀 Abandon and remove
lists	Okayish	-
lorem	Okayish	-
masters	Existing issues (#1386,#385) but okayish	-
math (*)	Ongoing work in progress	✏️ Continue work but some contributors would be great...
pagebuilder-bestfit (!!)	Non-general (#2327)	💀 Abandon and remove
pandoc	?	(I won't fight again over this one, but I stand by my older views...)
parallel	?	(I don't have an opinion on this one)
pdf	Ill-named	✏️ Should be "annotations", not "pdf" (and other related refactors)
pdfstructure	Not working well and not sane (#110)	✏️ Would need a study and redesign...
pullquote	Okayish	-
raiselower	Okayish	-
rebox	Okayish	-
retrograde (!!)	-	(I don't have an opinion on this one)
rotate	Existing issues and missing features	✏️ Consider fixes and improvements as per issues...
ruby	?	(I don't have an opinion on this one)
rules	Okayish	-
scalebox	Okayish	-
simpletable	Bad and too limited (#1206)	💀 Abandon and remove
specimen	?	(I don't have an opinion on this one)
svg	Issues (#1449,#1375,#1197 + no text support)	✏️ Consider fixes and improvements as per issues...
tableofcontents (!)	Okayish	-
tate	?	(I don't have an opinion on this one)
textcase	Okayish	-
twoside	Existing issues (#683) but okayish	-
unichar	Okayish	-
url	Okayish	-
verbatim (!)	Somewhat messy but okayish	-
xmltricks	?	(A mixed bad of things, not assessed here...)

Legend:

• (!) For the record, in my re·sil·ient collection, I use my own derivative versions of these packages.
• (!!) For the record, in my re·sil·ient collection, the use of these packages is forbidden.
• (*) Motivated contributors, feel free to ask what to help with...

[jodros]

I think ruby, hanmenkyoshi and tate should be gathered into a single cjk package.

For now, the classes could be reduced to base only, classes for CJK are a special case, and should be treated in the context of the Overhaul CJK.

A rewriting is worthwhile for many of those broken or messy packages, like frametricks, insertions, grid, bidi and so on...

Regarding all the other packages, I agree with your assessments.