summaryrefslogtreecommitdiff
path: root/src/client/textanalysis.js
diff options
context:
space:
mode:
authorAlon Levy <alon@pobox.com>2015-02-11 14:45:41 +0200
committerAlon Levy <alon@pobox.com>2015-02-11 14:45:41 +0200
commit6746a8fd949c401e5b2c1b00f3a7f69a3c209f36 (patch)
tree1d31b27674c3c5694160f280375b8cbb70be7f1b /src/client/textanalysis.js
parent0dc89aa6b16412b02d76151a777c2346409bb78f (diff)
client/textanalysis: sanitize quotes, fixes #293
Diffstat (limited to 'src/client/textanalysis.js')
-rw-r--r--src/client/textanalysis.js13
1 files changed, 12 insertions, 1 deletions
diff --git a/src/client/textanalysis.js b/src/client/textanalysis.js
index e0b57dbc..9a4cd412 100644
--- a/src/client/textanalysis.js
+++ b/src/client/textanalysis.js
@@ -28,6 +28,17 @@ function selectedType()
return nodetypes[typeindex];
}
+/*
+ * Place to do general cleanup of the text. Right now just replacing all quotation types
+ * with a single type, separately for double quotes and single quotes.
+ */
+function cleanup(text)
+{
+ return text
+ .replace(/[״”“„‟″‶]/, '"') // Convert typographic double quotes
+ .replace(/[`׳’‘‚‛′‵]/, "'"); // Convert typographic single quotes
+}
+
/**
* Tokenizer for input.
*
@@ -251,7 +262,7 @@ var textAnalyser = function (spec) {
//Sentence Sequencing
//Build the words and cuts the main elements
- tokens = tokenize(newtext, '#', '"');
+ tokens = tokenize(cleanup(newtext), '#', '"');
sentence = tokens.map(function (d) { return d.token; });
// build new node,link arrays in order of appearance