TeX - LaTeX Stack Exchange is a question and answer site for users of TeX, LaTeX, ConTeXt, and related typesetting systems. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

What are the various indexing packages and what are their pros and cons? When should one choose one over the other?

There is some discussion of this in What are the latest/best practices for index generation?, but I would prefer to have more details about each package.

share|improve this question
Sounds like a case for CW and one package per answer? What would be relevant categories of information to be included in each answer? – doncherry Apr 11 '12 at 21:18
@doncherry I'd rather have indexing specialists answer this in a single pass and gain reputation for their answer. – lockstep Apr 11 '12 at 21:20
up vote 17 down vote accepted

The features provided by the various indexing packages may be divided into three categories:

  1. Supporting multiple indexes;

  2. Providing macros that simplify index generation;

  3. Allowing for customization of the index layout.

The Topic Index of the TeX Catalogue lists about a dozen index-related packages; this overview is about the packages I've used (or at least read the manual).

makeidx, a "[s]tandard package for creating indexes", provides some index-related macros like \printindex which (for whatever reason) are not part of the LaTeX2e kernel (like \makeindex and \index) or the standard document classes (like the theindex environment).

multind "redefines [the] \makeindex, \index and \printindex commands to deal with multiple indexes". The package predates LaTeX2e; it doesn't work with the AMS classes, and index headings won't be formatted corresponding to other chapter/section headings.

index "supports multiple indexes in a single document and provides a more robust \index command". The package also offers \index* which typesets a word and creates an index entry for it as well. index is likely to produce incorrectly formatted index headings for non-standard classes.

splitidx/splitindex allow "more than 16 indexes: splitidx outputs only a single file \jobname.idx and the program [splitindex] splits that file into several raw index files and calls your favorite index processor for each of the files". The splitidx package offers the \extendtheindex macro that adds various hooks to the theindex environment; the actual customization of the latter is left to the user.

imakeidx "enables the user to produce and typeset one or more indexes simultaneously with a document" by using the shell-escape feature of modern TeX distributions. imakeidx is able to call the splitindex program (see above), and it features a key--value interface to customize various aspects of the index layout (including the possibility to call the MakeIndex program with a certain .ist style file). Originally, imakeidx was not compatible with the memoir class (which uses an idiosyncratic method to generate indexes), but v1.1 of the package does away with that restriction.

hvindex "simplifies the indexing of words [...] [T]o index a word in a text, you only have to type it once; the package makes sure it is both typeset and indexed". The package offers a bunch of macros that control the formatting of index entries; the package manual is silent about whether those macros may be used for the creation of multiple indexes.

idxlayout "offers a key--value interface to configure index layout parameters [and] is responsive to the index-related options and commands of the KOMA-Script and memoir classes". idxlayout does only deal with the theindex environment, not with .ist style files; it may be used on top of index, splitidx, and imakeidx. (Disclosure: I'm the package author.)

repeatindex "repeats [the] item of an index if a page or column break occurs within a list of subitems". The package README (from 2001) states that repeatindex "has not been tested much, yet" and lists some known problems.

share|improve this answer
I think one import point is the processor. I like xindy. – Marco Daniel May 7 '12 at 16:30
@MarcoDaniel I'll see if I can come up with something in the next days -- but comparing MakeIndex and xindy could be the topic of another answer, too. – lockstep May 7 '12 at 16:34
The links are broken; there seems to be a bug on CTAN's website: the /index part of the url leads to an index of all packages – Clément Feb 21 '15 at 16:41

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.