s434730/RandomSec
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Update TextFormatGuesser to support wikitext

Browse Source
This commit is contained in:
Antonin Delpeuch 2017-08-14 15:58:27 +01:00
parent e168c900e8
commit 73f7fdc036
1 changed files with 7 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
main/src/com/google/refine/importers/TextFormatGuesser.java
View File

@ -24,6 +24,8 @@ public class TextFormatGuesser implements FormatGuesser {
int closeBraces = 0;
int openAngleBrackets = 0;
int closeAngleBrackets = 0;
int wikiTableBegin = 0;
int wikiTableRow = 0;
int trailingPeriods = 0;
char firstChar = ' ';
@ -37,6 +39,8 @@ public class TextFormatGuesser implements FormatGuesser {
closeBraces += countSubstrings(chunk, "}");
openAngleBrackets += countSubstrings(chunk, "<");
closeAngleBrackets += countSubstrings(chunk, ">");
wikiTableBegin += countSubstrings(chunk, "{|");
wikiTableRow += countSubstrings(chunk, "|-");
trailingPeriods += countLineSuffix(chunk, ".");
if (!foundFirstChar) {
@ -50,7 +54,9 @@ public class TextFormatGuesser implements FormatGuesser {
}
if (foundFirstChar) {
if ((firstChar == '{' || firstChar == '[') &&
if (wikiTableBegin >= 1 && wikiTableRow >= 2) {
return "text/wiki";
} if ((firstChar == '{' || firstChar == '[') &&
openBraces >= 5 && closeBraces >= 5) {
return "text/json";
} else if (openAngleBrackets >= 5 && closeAngleBrackets >= 5) {