AUTO-MULTIPLE-CHOI(1)

📦 version 1.7.0, in arch rolling, nixpkgs unstable

🌸 名前 (NAME) 🌸

Okay so this is the manual page for auto-multiple-choice-annotate — yes, the AUTO-MULTIPLE-CHOI(1) one — and it’s literally the AMC tool that writes grading results onto multiple-choice exam answer sheets. It’s giving “teacher’s pet but make it Unix.” 💅

📝 概要 (SYNOPSIS) 📝

The vibe-check syntax is:

auto-multiple-choice annotate --project project-dir
                            --names-file students.csv [annotation options...]

✨ 解説 (DESCRIPTION) ✨

The auto-multiple-choice annotate command writes each question’s score and the overall score onto answer sheets, then outputs PDF files — either multiple files, one per candidate, or one single combined file. Periodt. 🎀

💅 全般オプション (General options) 💅

--project project-dir — sets the project name or directory.
--data project-data-dir — points to the directory where data files live; default is project-dir/data.
--pdf-dir pdf-dir — sets the project’s pdf directory; default is project-dir/cr/corrections/pdf.
--names-file students-list.csv — sets the candidate list filename.
--names-encoding encoding — picks the encoding of students-list.csv; default is utf-8.
--association-key key — sets the column used for candidate identification (a column name inside the candidate list file). The value actually used during identification is stored in the database and becomes the default value, so you should not use this option. 💀
--subject subject.pdf — sets the path to the exam question file; default is project-dir/DOC-sujet.pdf.
--compose mode — use this when bringing pages without boxes to mark (for example, problem-only pages before the answer sheet) over from the question paper. When mode is 1, pages are taken from the question paper, marks are written the same way as on the candidate’s answer sheet, and the same grading symbols are written onto the answer sheet. When mode is 2, pages are taken from the corrected answer sheet and grading symbols are not written (because it would duplicate them).
--corrected corrected.pdf — sets the path to the PDF model answer; default is project-dir/DOC-corrected.pdf. If the file does not exist, it is created using the values of the --n-copies, --with, --filter, and --filtered-source options.
--darkness-threshold t — sets the threshold for black-pixel ratio. The value actually used during grading is stored in the database and becomes the default value, so you should not use this option. 🙈
--darkness-threshold-up upper_t — sets the upper threshold for black-pixel ratio. The value actually used during grading is stored in the database and becomes the default value, so you should not use this option. 🙈
--csv-build-name name-model — sets the model for building candidate names from the candidate list file. The default value is (nom|surname) (prenom|name). In that case, the name is the contents of the nom column (or surname if nom is absent) followed by the contents of the prenom column (or name if prenom is absent), so simple French or English CSV files usually work.
--id-file ids.txt — specifies a file listing the IDs of answer sheets to annotate, one per line: in individual answer-sheet mode, the candidate number; in copy answer-sheet mode, student:copy. If no file is given, every answer sheet gets annotated.
--debug file.log — specifies the file where debug info gets logged.

📄 PDF出力オプション (PDF output options) 📄

--single-output — use this to combine annotated answer sheets into a single PDF file. The default behavior is to create one PDF file per candidate.
--sort s — sets the following sort keys; only effective when --single-output is used:
- l — candidate list file order.
- m — grade order.
- i — exam question ID order.
- n — name order.
--filename-model model — sets the filename model for annotated PDFs. In this model, specific sequences are replaced as follows:
- (N) — replaced by the candidate name (see --csv-build-name).
- (ID) — replaced by the candidate number.
- (COL) — replaced by the value of the COL column in the candidate list.

The default value is (N)-(ID).pdf.

🧚 ソースファイルオプション (Source file options) 🧚

The following options are used when the file given by --corrected does not exist and needs to be rebuilt:

--src mcq-source-file --n-copies n --with latex-engine --filter
filter-name --filtered-source fs.tex

See auto-multiple-choice-prepare(1). 🫶

🦋 スキャン画像埋込み用オプション (Scanned-image embedding options) 🦋

If scanned images exist, they get embedded into the annotated PDF file. These options control the quality of the embedded scanned images and let you shrink the file size:

--embedded-max-size maxsize — sets the maximum size of scanned images in pixels, in the form widthxheight. The default value is empty, meaning no upper limit. Scanned images larger than the specified size are shrunk before being embedded in the annotated file.
--embedded-format format — sets the embedding format to jpeg or png.
--embedded-jpeg-quality quality — sets the JPEG quality, from 1 to 100, when embedding as JPEG.

🌈 採点記入オプション (Annotation options) 🌈

These options control what kind of grading gets written:

--font-size fs — sets the font size in points.
--text-color color — sets the text color. The default value is red.
--verdict header — sets the text header shown on the first page of each candidate’s answer sheet. This text can contain newline characters for a multi-line header. These specific sequences get replaced:
- %S — replaced by the candidate’s raw-score total.
- %M — replaced by the maximum raw-score total.
- %s — replaced by the candidate’s score.
- %m — replaced by the maximum score.
- %(ID) — replaced by the candidate name.
- %(COL) — replaced by the value of the COL column in the candidate list.
--position pos — sets where each question’s grading gets written:
- marge — write in the left margin.
- marges — write in the nearest margin (left or right).
- case — write near the box to be marked.
- none — do not write grading.
--verdict-question perl-expression — sets the text printed next to each question as an expression evaluated by Perl (if you need plain text, wrap it in "). Some sequences are replaced before evaluation:
- %S — replaced by this question’s raw score.
- %M — replaced by this question’s maximum raw score.
- %s — same as %S, but rounded to nc digits (see the --ch-sign option).
- %m — same as %M, but rounded to nc digits (see the --ch-sign option).

The default value for this option is "%s/%m". You can also write conditional expressions using the Perl syntax ( TEST ? IF-YES : IF-NO ). To write pass if full marks, otherwise fail, do it like this:

--verdict-question "(%S==%M ? \"合格\" : \"不合格\")"

--verdict-question-cancelled perl-expression — same as --verdict-question, but applied to cancelled questions (see the allowempty grading criterion). The default value is "X".
--indicatives — use this option to write grading on non-graded questions too. (These questions’ raw scores are not counted when computing the overall raw score. Correct/incorrect status is usually irrelevant for these questions, so no correct/incorrect marks are written in the corresponding boxes.)
--line-width width — sets the line width, in pixels, of symbols written during grading. The default value is 2.
--symbols symbols-definition — defines how boxes get symbols. symbols-definition is a comma-separated list in the form H-A:shape:color, where H is 0 or 1 and says whether the box should not be marked or should be marked, A is 0 or 1 and says whether the box is not marked or is marked, shape is one of none, circle, mark, or box, and color is a color (a name or #RGB; see the ImageMagick or GraphicsMagick documentation for details). The default value is 0-0:none,0-1:circle:red,1-0:mark:red,1-1:mark:blue.

📏 長さ (Lengths) 📏

You can specify units for these lengths: in, ft, pt, cm, mm.

--dist-to-box d — when using --position case, sets the distance of the annotation position from the box. Default is 1cm.
--dist-margin d — sets the margin size. Default is 5mm.
--dist-margin-global d — sets the header margin size. Default is 3mm.

👑 著者 (AUTHORS) 👑

Bienvenue Alexis[FAMILY Given] <paamc@passoire.fr> — main author 💖
Sarkar Anirvan[FAMILY Given] — author and editor
Kagotani Hiroto[FAMILY Given] — editor
Bréal Frédéric[FAMILY Given] — author and editor
Bérard Jean[FAMILY Given] — translator from the French version to the English version
Khaznadar Georges[FAMILY Given] — translator from the French version to the English version
Kagotani Hiroto[FAMILY Given] — translator from the English version to the Japanese version

🌟 製作著作 (COPYRIGHT) 🌟

This document can be used under the GNU General Public License version 2 or later.

💖 That’s a wrap, bestie — brought to you by Auto Multiple Choice 1.7.0, dated 2025-04-17. Slay! 💅✨

📦 where this page comes from

arch rolling: auto-multiple-choice 1.7.0-7 · GPL-2.0-or-later
nixpkgs unstable: auto-multiple-choice 1.7.0
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on 2026-11-04: a language model's rewrite of the original, under its terms