5 files changed, 59 insertions, 8 deletions
diff --git a/backend/data/README.md b/backend/data/README.md
index 88053b5..b4ee77e 100644
--- a/backend/data/README.md
+++ b/backend/data/README.md
@@ -5,9 +5,26 @@ File Generation Process
     table using data in otol/*.
 3   Run genEolNameData.py, which adds a 'names' table to data.db, 
     using data in eol/vernacularNames.csv and the 'nodes' table.
-4   Use downloadImgsForReview.py to download EOL images into imgsForReview/.
+4   Run genSpellfixNameData.py, which adds a 'spellfix\_alt\_names'
+    table to data.db, using data in the 'names' table.
+5   Use downloadImgsForReview.py to download EOL images into imgsForReview/.
     It uses data in eol/imagesList.db, and the 'nodes' table.
-5   Use reviewImgs.py to filter images in imgsForReview/ into EOL-id-unique
+6   Use reviewImgs.py to filter images in imgsForReview/ into EOL-id-unique
     images in imgsReviewed/ (uses 'names' to display common names).
-6   Use genImgsForWeb.py to create cropped/resized images in img/, using
+7   Use genImgsForWeb.py to create cropped/resized images in img/, using
     images in imgsReviewed, and also to add an 'images' table to data.db.
+
+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/genEolNameData.py b/backend/data/genEolNameData.py
index 5070fd7..46e109a 100755
--- a/backend/data/genEolNameData.py
+++ b/backend/data/genEolNameData.py
@@ -1,7 +1,5 @@
 #!/usr/bin/python3
 
-# 
-
 import sys, re
 import csv, sqlite3
 
@@ -13,6 +11,9 @@ usageInfo += "\n"
 usageInfo += "Expects a CSV header describing lines with format:\n"
 usageInfo += "    page_id, canonical_form, vernacular_string, language_code,\n"
 usageInfo += "    resource_name, is_preferred_by_resource, is_preferred_by_eol\n"
+if len(sys.argv) > 1:
+	print(usageInfo, file=sys.stderr)
+	sys.exit(1)
 
 vnamesFile = "eol/vernacularNames.csv"
 dbFile = "data.db"
@@ -54,6 +55,7 @@ with open(vnamesFile, newline="") as csvfile:
 		# Add to maps
 		updateMaps(name1, pid, True, False)
 		updateMaps(name2, pid, False, preferred)
+
 # Open db connection
 dbCon = sqlite3.connect(dbFile)
 dbCur = dbCon.cursor()
@@ -62,9 +64,9 @@ dbCur.execute("CREATE TABLE names(name TEXT, alt_name TEXT, eol_id INT, pref_alt
 # Iterate through 'nodes' table, resolving to canonical-names
 usedPids = set()
 unresolvedNodeNames = set()
-cur2 = dbCon.cursor()
+dbCur2 = dbCon.cursor()
 iterationNum = 0
-for row in cur2.execute("SELECT name FROM nodes"):
+for row in dbCur2.execute("SELECT name FROM nodes"):
 	name = row[0]
 	iterationNum += 1
 	if iterationNum % 10000 == 0:
@@ -108,5 +110,6 @@ for name in unresolvedNodeNames:
 		for n in altNames:
 			isPreferred = 1 if (n == preferredName) else 0
 			dbCur.execute("INSERT INTO names VALUES (?, ?, ?, ?)", (name, n, pidToUse, isPreferred))
+# Close db
 dbCon.commit()
 dbCon.close()
diff --git a/backend/data/genOtolData.py b/backend/data/genOtolData.py
index 696d59b..57a15d2 100755
--- a/backend/data/genOtolData.py
+++ b/backend/data/genOtolData.py
@@ -18,7 +18,6 @@ usageInfo += "Expected annotations.json format:\n"
 usageInfo += "    JSON object holding information about the tree-of-life release.\n"
 usageInfo += "    The object's 'nodes' field maps node IDs to objects holding information about that node,\n"
 usageInfo += "    such as phylogenetic trees that support/conflict with it's placement.\n"
-
 if len(sys.argv) > 1:
 	print(usageInfo, file=sys.stderr)
 	sys.exit(1)
diff --git a/backend/data/genSpellfixNameData.py b/backend/data/genSpellfixNameData.py
new file mode 100755
index 0000000..9a3a7a1
--- /dev/null
+++ b/backend/data/genSpellfixNameData.py
@@ -0,0 +1,32 @@
+#!/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
new file mode 100755
index 0000000..0bc985c
--- /dev/null
+++ b/backend/data/spellfix.so