cbJisho Usage

Table of Contents

Description

[top]

cbJisho is an English-to-Japanese dictionary that sorts results based on relative frequencies in blogs, newspapers and novels. You can filter words by JLPT level (1-5), whether or not it’s a JDIC "common" word and by character length of the definition. You can also search using a regular expression or by whole word only. Search text can either be in English (in which case the definitions are searched) or Japanese (in which case the kanji and kana fields are searched) or you can also use a simple SQL query. Results can be saved to the clipboard or to a file.

If you want to change the default settings, you may edit settings.txt.

If you want to change the colors/fonts/styles of the results window, you may edit Template/results_template.html.

If you want to remove or re-order the fields of the results window, you may edit Tempate/single_result_template.html.

The Interface

[top]






Keyboard Shortcuts

[top]

Key Description
ENTER Peform search
ESC Clear the search box
CTRL-L Send focus to the search box
UP Move back through the search history
DOWN Move forward through the search history
CTRL-D Copy the results to the clipboard. See Saving Results to File/Clipboard.
CTRL-S Copy the results to file. See Saving Results to File/Clipboard.
CTRL-W Toggle the Whole word checkbox
CTRL-R Toggle the RegEx checkbox
CTRL-J Toggle all of the JLPT buttons
CTRL-P Toggle the (P) checkbox
CTRL-H Show this help page
CTRL-Q Show the SQL help page

Saving Results to File/Clipboard

[top]

Save to the Clipboard:

Press CTRL-D.

Save to a File:

  1. Open settings.txt (it's in the same directory as cbJisho.exe).
  2. Find the "SaveFile" setting and set it to the full path of the file that you want to save to.

    Example:
    SaveFile = C:\Temp\results.txt

  3. Open cbJisho.
  4. Press CTRL-S to save the results to file.

Formatting the saved results:

You can specify the format of the saved results by editing the "SaveFormat" setting in settings.txt with the following tokens:

Token Description
$s Sequence Number
$o Overall Frequency
$b Blog Frequency
$n Newspaper Frequency
$v Novel Frequency
$j JLPT Number
$p (P)
$k Kanji
$a Kana
$d Defination

Example:
To save the overall frequency, the kanji, the kana and the definition, use the following:
SaveFormat = $o$k$a$d

Notes:

  1. The fields will be automatically seperated with a tab character.
  2. All results are copied, not just the results that are displayed.
  3. If you don't specify a file, CTRL-S will save to the clipboard instead.

SQL Query Help

[top]

How to use basic SQL queries with cbJisho:

Start the search with a forward slash ("/"). The following will automatically be prepended to the query (so don't type it):  SELECT * FROM Dict WHERE

You may use the following columns:

Column Type Range Description
kanji TEXT n/a The kanji of an entry
kana TEXT n/a The kana of an entry
def TEXT n/a The description of an entry
overall REAL [0-100] The overall frequency of the entry (based on blog, newspaper and novel frequencies)
blog REAL [0-100] The blog frequency. It is set to -1 if no frequency is associated with the entry.
news REAL [0-100] The newspaper frequency. It is set to -1 if no frequency is associated with the entry.
novel REAL [0-100] The novel frequency. It is set to -1 if no frequency is associated with the entry.
jlpt INTEGER [1-5] The JLPT level
common INTEGER [0-1] 1 = Entry is EDICT common "(P)" word

Note: The "Whole Word" and "RegEx" options have no effect when using SQL.

Examples:

  1. Find entries where the kanji starts with "時":
    /kanji like '時%'

  2. Find entries where the kanji is 3 characters and the middle character is "時":
    /kanji like '_時_'

  3. Find entries where the definition contains "candy" and the kanji contains "子":
    /def like '%candy%' and kanji like '%子%'

  4. Find godan verbs for "love" or "hate":
    /def like '%v5%' and (def like '%love%' or def like '%hate%')

  5. Definition contains "tiger" and the kana has fewer than 4 characters:
    /def like '%tiger%' and length(kana) < 4

  6. Find entries where the novel frequency is at least 99.0:
    /novel >= 99.0

  7. Find entries where the definition contains "US" but not "us" or "Us":
    /def glob '*US*'

Most useful SQL operators to use with cbJisho:

Questions and Answers

[top]

Question:
Where did these frequencies come from?

Answer:
The blog and newspaper frequency lists where obtained from The Monash Nihongo ftp Archive.

Blog frequencies were obtained by using Google search result counts within Goo Blog. For details, see the readme.

Newspaper frequencies are based on a large number of articles from the online versions of the Yomiuri and Mainichi newspapers. For details, see the readme.

Novel frequencies are based on 5109 novels. The list of novels used: http://pastebin.com/VLJpTREd. The first 50 lines and last 20 lines were removed from each file so that things like table of contents, copyright and publisher information were not parsed. The readings (between 《 and 》) were also removed.

Overall frequencies are an average of the blog, newspaper and novel frequencies.