diff options
Diffstat (limited to 'backend/data')
| -rw-r--r-- | backend/data/README.md | 19 | ||||
| -rwxr-xr-x | backend/data/genSpellfixNameData.py | 32 | ||||
| -rwxr-xr-x | backend/data/spellfix.so | bin | 86456 -> 0 bytes |
3 files changed, 0 insertions, 51 deletions
diff --git a/backend/data/README.md b/backend/data/README.md index 9f0ea82..27619de 100644 --- a/backend/data/README.md +++ b/backend/data/README.md @@ -9,8 +9,6 @@ File Generation Process 1 Obtain data in eol/, as specified in it's README. 2 Run genEolNameData.py, which adds 'names' and 'eol\_ids' tables to data.db, using data in eol/vernacularNames.csv and the 'nodes' table. - 3 Run genSpellfixNameData.py, which adds a 'spellfix\_alt\_names' - table to data.db, using data in the 'names' table. 3 Image Data 1 Use downloadImgsForReview.py to download EOL images into imgsForReview/. It uses data in eol/imagesList.db, and the 'eol\_ids' table. @@ -31,24 +29,7 @@ data.db tables name TEXT, alt\_name TEXT, pref\_alt INT, PRIMARY KEY(name, alt\_name) - eol\_ids <br> id INT PRIMARY KEY, name TEXT -- spellfix\_alt\_names - images <br> eol\_id INT PRIMARY KEY, source\_url TEXT, license TEXT, copyright\_owner TEXT - descs <br> name TEXT PRIMARY KEY, desc TEXT, redirected INT - -spellfix.so -=========== - -This file provides the spellfix1 extension for Sqlite, and -is used for responding to fuzzy-search requests. - -It was obtained by: -1 Downloading the sqlite source tree from - the github mirror at <https://github.com/sqlite/sqlite>, - into a directory sqlite/ -2 After making sure autoconf 2.61+ and libtool are installed, - running `mkdir bld; cd bld; ../sqlite/configure;` -3 Running `make` -4 Running `cp ../sqlite/ext/misc/spellfix.c .` -5 Running `gcc -fPIC -shared spellfix.c -o spellfix.so` diff --git a/backend/data/genSpellfixNameData.py b/backend/data/genSpellfixNameData.py deleted file mode 100755 index 9a3a7a1..0000000 --- a/backend/data/genSpellfixNameData.py +++ /dev/null @@ -1,32 +0,0 @@ -#!/usr/bin/python3 - -import sys -import sqlite3 - -usageInfo = f"usage: {sys.argv[0]}\n" -usageInfo += "Reads alt-names from a 'names' table in a database, and adds a spellfix \n" -usageInfo += "table 'spellfix_alt_names' usable for fuzzy-searching those names.\n" -if len(sys.argv) > 1: - print(usageInfo, file=sys.stderr) - sys.exit(1) - -dbFile = "data.db" - -# Connect to db, and load spellfix extension -dbCon = sqlite3.connect(dbFile) -dbCon.enable_load_extension(True) -dbCon.load_extension('./spellfix') -# Create spellfix table, and insert alt-names -spellfixCur = dbCon.cursor() -spellfixCur.execute("CREATE VIRTUAL TABLE spellfix_alt_names USING spellfix1") -namesCur = dbCon.cursor() -iterationNum = 0 -for row in namesCur.execute("SELECT DISTINCT alt_name FROM names"): - iterationNum += 1 - if iterationNum % 10000 == 0: - print("Loop {}: {}".format(iterationNum, row[0])) - # Insert alt-name - spellfixCur.execute("INSERT INTO spellfix_alt_names(word) VALUES (?)", (row[0],)) -# Close db -dbCon.commit() -dbCon.close() diff --git a/backend/data/spellfix.so b/backend/data/spellfix.so Binary files differdeleted file mode 100755 index 0bc985c..0000000 --- a/backend/data/spellfix.so +++ /dev/null |
