Skip to content

Latest commit

 

History

History
310 lines (219 loc) · 7.14 KB

README.md

File metadata and controls

310 lines (219 loc) · 7.14 KB
Raw

pbjkit

Command-line Chinese-English dictionary development kit for Pinyinbase, CC-CEDICT and more.

Tools in pbjkit

NOTE: pbj is baking cakes from ingredients. pbjson is putting baked cakes into JSON packages.

INSTALL

Requirements

From within your home (~) directory, install Pinyinbase:

$ npm i -g git+https://github.com/pffy/pbjkit.git

Alternatively, you can can install Pinyinbase like this:

$ npm i -g github:pffy/pbjkit

After installation, make sure everything is peachy keen:

$ pbj

You are all set!

TROUBLESHOOT

On Linus for Chromebooks or Virtual Machines (VMs) you may have restricted user accounts. This may prevent the global installation of npm in the default locations.

This issue may be called an EACCES error or something similar.

To fix this issue, first make a new local npm global folder in your home directory.

$ mkdir ~/.npm-global

Then, tell npm to use that new global folder:

$ npm config set prefix '~/.npm-global'

If you have a ~/.profile file in your home directory, then add this line:

export PATH=~/.npm-global/bin:$PATH

If there is no ~/profile file, you can simply create a new file:

$ echo export PATH=~/.npm-global/bin:$PATH > ~/.profile

Then, on the command line, enter this:

$ source ~/.profile

Now, try to reinstall:

$ npm i -g github:pffy/pbjkit

Then, check your install:

$ pbj

NOTE: Details about this and similar fixes for the EACCES error issue on Linux for Chromebook.

QUICK START

1. Download Pinyinbase or any other distributed CEDICT-format dictionary into your working directory:

$ git clone https://github.com/pffy/pinyinbase

2. Combine all the glossaries in the folder into a single CEDICT file:

$ pbj -i ./pinyinbase/vocab/*.txt

The output file is pb.txt (a CEDICT-format dictionary file).

pbj

Overview

  • Command-line Pinyinbase Joiner utility
  • Combine multiple Pinyinase glossaries into a single data source file

Options

--version

Displays the version, then exits:

$ pbj --version

--help

Displays the help information, then exits:

$ pbj --help

-i

Input Pinyinbase glossary files to be validated, processed, and combined.

$ pbj -i file1 file2 file3

pbj performs validation:

  • Each file must be a valid Pinyinbase glossary.
  • Each line of a valid Pinyinbase glossary must be a CEDICT-formatted dictionary entry.

All three files will be combined into a single CEDICT-formatted file called pb.txt.

-d

Adds epoch date to output filename.

$ pbj -d -i file1 file2

Output file is similar to pb-1605898061.txt.

-q

Quiet processing. 🤐

$ pbj -q -i ./folder/*.txt

Valid Pinyinbase glossaries

NOTE: For examples of glossary files, please visit the Pinyinbase repo.

  • The filename must exactly match the first comment line to be a valid glossary file.

    • Otherwise, the file will be ignored by pbj.

  • EXAMPLE: file1.txt

    • The following is a blank AND perfectly valid glossary file.
    • Notice there are no entries in this file.
# file1.txt
  • EXAMPLE: file2.txt
    • The following glossary is invalid. The filenames and first line comment do not exactly match.
# file3.txt
  • EXAMPLE: file8.txt
    • The following glossary is valid. The filenames and first line comment do exactly match.
    • There are no valid CEDICT-formatted entires in this file.
# file8.txt
some text here whatever
more text here whatever
this is not not valid text

pbjson

Overview

  • Command-line Pinyinbase JSON document generator
  • Convert a compiled Chinese-English dictionary into a JSON file
  • Adds metapinyin data for better entry storage and retrieval

--version

Shows version, then exits.

$ pbjson --version

--help

Shows help, then exits.

$ pbjson --help

-i

Input CEDICT-formatted dictionary files to be validated, processed, and combined.

$ pbjson -i file1 file2 file3

pbjson performs validation:

  • Each line of the input dictionary file must be a CEDICT-formatted dictionary entry; otherwise, the line is ignored.

All three files will be combined into a single JSON document called pb.json.

-d

Adds epoch date to output file.

$ pbjson -d -i file.*txt
$ pbjson -d -i cedict1 cedict2 ./folder/*.txt

Output file is similar to pb-1605898061.json.

-q

Quiet processing. 🤐

$ pbjson -q -i ./folder/*.txt

Making a custom CEDICT dictionary using the xargs command

  • In a working directory containing Pinyinbase glossaries.
$ cd ./pinyinbase/vocab
  • With a custom list of glossaries in a text file called astro1.txt:
vocab-cmn-astronomy-planets-earth-moons.txt
vocab-cmn-astronomy-planets-jupiter-moons.txt
vocab-cmn-astronomy-planets-mars-moons.txt
vocab-cmn-astronomy-planets.txt
vocab-cmn-astronomy-thesun.txt

NOTE: The text file astro1.txt should only contains glossary file names -- not CEDICT-formatted entries.

For example, in the pinyinbase/vocab folder containing a new file called astro1.txt:

parent/
  - pinyinbase/
    - vocab/  <-- YOU ARE HERE
      - astro1.txt
      - ...

$ xargs pbj -i < astro1.txt

Only the glossaries in the file astro1.txt are processed. The glossaries are combined into a single file called pb.txt.

This command pattern also works for pbjson.

PBJ vs PBJSON: pbj validates Pinyinbase glossaries and CEDICT syntax and combines files.pbjson only validates CEDICT syntax and converts CEDICT files into JSON documents.

For example, in the pinyinbase/vocab folder, where the file astro2.txt is in the parent folder:

parent/
  - astro2.txt
  - pinyinbase/
    - vocab/ <-- YOU ARE HERE
    - ...

$ xargs pbjson -i < ../../astro2.txt

Only the dictionary files in the file astro2.txt are processed. The dictionary files are combined into a single file called pb.json.

Alternative to xargs

As alternative to xargs, you can use cat.

In this example, the file astro2.txt is the parent directory along with pinyinbase folder:

parent/ <-- YOU ARE HERE
  - astro2.txt
  - pinyinbase/
    - vocab/
    - ...

We can extract the file list for pbjson using cat as follows:

$ pbjson -i ./pinyinbase/vocab/$(cat astro2.txt)

As you can see, you can build several custom CEDICT dictionaries very quickly by simply using dictionary file lists as recipes.

Licenses