MENDEX(1)

🎀 MENDEX(1) 🎀

MENDEX(1) from the General Commands Manual, bestie, and the star of this page is mendex 💅

✨ NAME ✨

mendex - Japanese index processor 🇯🇵

✨ SYNOPSIS ✨

mendex [-ilqrcgfEJSU] [-s sty] [-d dic] [-o ind] [-t log] [-p no] [-I
enc] [--help] [--] [ idx0 idx1 idx2 ...]

✨ DESCRIPTION ✨

okay so mendex is a general purpose hierarchical index generator 🧚 It takes one or more input files (.idx, which a text formatter like LaTeX usually whips up for you), sorts all the entries, and spits out an output file that's ready to be formatted. It's almost compatible with makeindex, and it also has an extra feature for handling the readings of kanji words, which is so iconic 💜

The input and output formats are set up in a style file. The readings of kanji words go in a dictionary file. Your index can nest up to three levels deep (0, 1, and 2) of subitems, so you can get really organized 🌸

✨ OPTIONS ✨

• -i — Reads from stdin even when you've given it index files. Main character energy 📥
• -l — Switches to ´sort by character order´. By default you get ´sort by word order´ instead, and I'll explain the difference below 🦋
• -q — Quiet mode 🤫 It won't send any messages to stderr, except for error messages and warnings (those always get through, periodt)
• -r — Turns off implicit page range formation. By default, three or more pages in a row get abbreviated into a range, like 1–5 ✨
• -c — Squishes runs of in-between blanks (spaces and/or tabs) into one space, and drops any blanks at the start and end. By default, the blanks in your index key stay exactly as they are 🙈
• -g — Makes the Japanese index head A-line (A, Ka, Sa, ...; 10 characters) from the gojuon table (the Japanese syllabary). By default, all 48 characters in the gojuon table get used 🌈
• -f — Forces kanji output even when they don't exist in the dictionaries. Bold choice 🔥
• -s sty — Uses sty as the style file 🎀
• -d dic — Uses dic as the dictionary file. The dictionary is made of lists of <index_word reading> pairs 📖
• -o ind — Uses ind as the output index file. By default, the name comes from the base name of the first input file with the extension ind tacked on 🍓
• -t log — Uses log as the transcript file. By default, the name comes from the base name of the first input file with the extension ilg tacked on 📝
• -p no — Sets the starting page number of the output index list to no. The argument can be a number, or one of these: any (the next page to the end of contents), odd (the next odd page to the end of contents), or even (the next even page to the end of contents) 🦄
• -E — Sets input/output character encoding to EUC-JP 🎐
• -J — Sets input/output character encoding to JIS (ISO-2022-JP) 🌸
• -S — Sets input/output character encoding to Shift_JIS 💖
• -U — Sets input/output character encoding to UTF-8 🫶
• -I enc — Sets the internal character encoding to enc. The argument can be euc (EUC-JP) or utf8 (UTF-8). It defaults to utf8, so you're set unless you say otherwise 💅
• --help — Shows a summary of the options 🧚
• -- — Anything after -- won't be treated as an option. This is a lifesaver when your input file name starts with - 🙈

✨ STYLE FILE ✨

okay this part is the blueprint bestie 📐 The style file tells mendex what the .idx input files look like, and what the final output file should look like. The format is upper compatible with makeindex's, so if you know that one you're halfway there ✨ The file is a list of <specifier attribute> pairs. There are two kinds of specifiers, input and output, and the pairs can show up in any order you like. A line that starts with ´%´ is a comment, so you can leave yourself notes 💬

✨ Input file style parameter ✨

• keyword <string> (default "\\indexentry") — The command with an argument that holds the index entry being processed 📮
• arg_open <char> (default ´{´) — The opening delimiter that marks where an index entry starts 🎀
• arg_close <char> (default ´}´) — The closing delimiter that marks where an index entry ends 🎀
• range_open <char> (default ´(´) — The opening delimiter that marks where a page range starts 🦋
• range_close <char> (default ´)´) — The closing delimiter that marks where a page range ends 🦋
• level <char> (default ´!´) — The delimiter that marks a lower level 🪜
• actual <char> (default ´@´) — The symbol saying the next sequence should show up as the index strings in the output file 💜
• encap <char> (default ´|´) — The symbol saying the next sequence is a command name to attach to the page number ✨
• page_compositor <string> (default "-") — The separator between page levels, for styles that have multi-level page numbers 🌈
• page_precedence <string> (default "rnaRA") — The priority order for expressing page numbers. R and r mean Roman, n means arabic numeral, and A and a mean Latin alphabet 🔥
• quote <char> (default ´"´) — The escape character for mendex parameters 🫧
• escape <char> (default ´\\´) — The escape character for general scripts 🙃

✨ Output file style parameter ✨

• preamble <string> (default "\\begin{theindex}\n") — The preamble of the output file 🌸
• postamble <string> (default "\n\n\\end{theindex}\n") — The postamble of the output file 🫶
• setpage_prefix <string> (default "\n \\setcounter{page}{") — The prefix for the page number when a start page is set 🍓
• setpage_suffix <string> (default "}\n") — The suffix for the page number when a start page is set 🍓
• group_skip <string> (default "\n\n \\indexspace\n") — The strings that add vertical space before a new section of the index 🌈
• lethead_prefix <string> (default "") — The prefix for the heading of a newly appearing heading letter 💖
• heading_prefix <string> (default "") — Same as lethead_prefix, kept around for makeindex compatibility 🎐
• lethead_suffix <string> (default "") — The suffix for the heading of a newly appearing heading letter 💖
• heading_suffix <string> (default "") — Same as lethead_suffix, kept around for makeindex compatibility 🎐
• lethead_flag <number> (default 0) — The flag that controls whether heading letters get output. 0 means no output, 1 means uppercase, and -1 means lowercase 🔠
• heading_flag <number> (default 0) — Same as lethead_flag. Heads up: makeindex calls this one headings_flag instead 👀
• item_0 <string> (default "\n \\item ") — The command sequence inserted between primary level entries 🧁
• item_1 <string> (default "\n \\subitem ") — The command sequence inserted between sub level entries 🧁
• item_2 <string> (default "\n \\subsubitem ") — The command sequence inserted between subsub level entries 🧁
• item_01 <string> (default "\n \\subitem ") — The command sequence inserted between primary and sub level entries 🧁
• item_x1 <string> (default "\n \\subitem ") — The command sequence inserted between primary and sub level entries when the main entry doesn't have a page number 🥺
• item_12 <string> (default "\n \\subsubitem ") — The command sequence inserted between sub and subsub level entries 🧁
• item_x2 <string> (default "\n \\subsubitem ") — The command sequence inserted between sub and subsub level entries when the sub level entry doesn't have a page number 🥺
• encap_prefix <string> (default "\\") — the little prefix that goes in front of an encapsulating command when it gets stuck onto the page number 💅
• encap_infix <string> (default "{") — the bit that sits right before the page number when the encapsulating command is added 🎀
• encap_suffix <string> (default "}") — the closing bit that comes after the page number when the encapsulating command is added ✨
• line_max <number> (default 72) — the max length of one line 📏 If a line goes past this number, it gets folded 🦋
• indent_space <string> (default "") — the space used to indent the top of a folded line 🫶
• indent_length <number> (default 16) — how long that indent space is when it's inserted at the top of a folded line 🌸
• symhead_positive <string> (default "Symbols") — the heading letter for numbers and symbols when lethead_flag or heading_flag is a positive number 💜
• symhead_negative <string> (default "symbols") — the heading letter for numbers and symbols when lethead_flag or heading_flag is a negative number 🥀
• symbol <string> (default "") — the heading letter for numbers and symbols when symbol_flag is non zero 🔥 If you set this, it takes priority over symhead_positive and symhead_negative. (Extended by mendex)
• symbol_flag <number> (default 1) — the flag for outputting symbols 🙈 If it's 0, nothing gets output. (Extended by mendex)
• letter_head <number> (default 1) — the flag for the heading letter for Japanese Kana 🇯🇵 1 uses Katakana, 2 uses Hiragana. (Extended by mendex)
• priority <number> (default 0) — the flag for how index words made of Japanese and non-Japanese text (ex. Latin scripts) get sorted 🌈 If it's non zero, one space (U+20) gets inserted between a Japanese sequence and a non-Japanese sequence during sorting. (Extended by mendex)
• character_order <string> (default "SNEJ") — the order of scripts and symbols 👑 S, N, E and J stand for symbol, numbers, English and Japanese. (Extended by mendex) > Make sure S and N are right next to each other, because mendex treats numbers as part of the symbol group. Don't skip this one bestie 😭

---

✨ ABOUT JAPANESE PROCESSING ✨

okay so mendex has an extra feature that makes Japanese indexes way easier to handle than with makeindex 🎀 You don't have to manually type out a reading for every single kanji word. Lowkey iconic, no cap.

Japanese kanji words usually get sorted by the syllables of their readings (´Yomi´), and those readings can be written with kana (Hiragana or Katakana) 🌸 mendex can accept index words written straight in kana in your input file, and it can also convert index words into kana by checking Japanese dictionaries 📖

Here are some examples of how syllables get simplified internally 💅

かぶしきがいしゃ         かふしきかいしや
マッキントッシュ         まつきんとつしゆ
ワープロ            わあふろ

The dictionary file is a list of <´index_word´ ´reading´> pairs 🦋 The index word can be in any script (kanji, kana, etc), but the reading has to be in Hiragana or Katakana. The thing separating the index word and its reading is one or more tab(s) or space(s). Here's an example dictionary 🧚

漢字      かんじ
読み      よみ
環境      かんきょう
α         アルファ

Here's the tea ☕ each index word can only have one Yomi in the dictionary. Some kanji words (ex. 「表」) can have more than one Yomi (ex. 「ひょう」 and 「おもて」), but only one of them can be registered 🙈 If you need a different Yomi, you have to write it out explicitly in kana on the input file, like \index{ひょう@表} or \index{おもて@表} 💜

Also, a dictionary file gets read automatically if you set its file name in the environment variable INDEXDEFAULTDICTIONARY ✨ The dictionary set by that environment variable can be used together with the file(s) you pass with the -d option 🫶

---

✨ ABOUT SORTING PROCEDURE ✨

by default mendex sorts indexes as is (´sort by word order´) 📚 If you set the -l option, spaces between words in an index get truncated before the sorting happens (´sort by character order´) 🌈 Even when sorting by character order, the index in the output keeps its original sequence, without the truncation. Main character energy 💁‍♀️

Here's an example 🎀

sort by word order       sort by character order
X Window            Xlib
Xlib                XView
XView                    X Window

There are also two sorting methods for indexes that mix Japanese kana with other scripts (e.g. Latin script) 🇯🇵 In a style file, setting priority to 0 (the default) or 1 controls whether a space gets inserted between Japanese Kana and other scripts before the sorting happens. priority=0 inserts the space, and priority=1 does not 🦋 Here's how that looks ✨

priority=0               priority=1
index sort               indファイル
indファイル              index sort

---

✨ ENVIRONMENT VARIABLES ✨

mendex checks these environment variables 💜

• INDEXSTYLE — the directory where index style files live 📁
• INDEXDEFAULTSTYLE — the index style file that gets used as the default 🎀
• INDEXDICTIONARY — the directory where dictionary files live 📖
• INDEXDEFAULTDICTIONARY — the dictionary file that gets read automatically 🫶

---

✨ DETAIL ✨

the detailed specification is compatible with makeindex 🧚 So if you already know makeindex, you're basically already a pro 👑

---

✨ KNOWN ISSUES ✨

⚠️ bestie, heads up: when you use a plural page number expression, you should specify the .idx files along with the order of the page numbers. If you don't, wrong page numbers might get output 😭

---

✨ SEE ALSO ✨

tex(1), latex(1), makeindex(1) 🌸

---

✨ AUTHOR ✨

This manual page was written by Japanese TeX Development Community 💖

---

Okay that's the whole page, you made it bestie 🥹🫶 MENDEX(1) forever, periodt 💅✨