sjs/media
1
0
Fork 0
mirror of https://github.com/samsonjs/media.git synced 2026-04-27 15:07:40 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Modified Webvtt parser to handle cue identifiers and tags.

Browse source 
Issue: #268
This commit is contained in:
Oliver Woodman 2015-02-03 12:41:49 +00:00
parent c1a2f3d0c2
commit 0ef28abbec
1 changed files with 32 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

36
library/src/main/java/com/google/android/exoplayer/text/webvtt/WebvttParser.java
View file

@ -56,20 +56,28 @@ public class WebvttParser implements SubtitleParser {
private static final Pattern WEBVTT_METADATA_HEADER = private static final Pattern WEBVTT_METADATA_HEADER =
Pattern.compile(WEBVTT_METADATA_HEADER_STRING); Pattern.compile(WEBVTT_METADATA_HEADER_STRING);
private static final String WEBVTT_CUE_IDENTIFIER_STRING = "^(?!.*(-->)).*$";
private static final Pattern WEBVTT_CUE_IDENTIFIER =
Pattern.compile(WEBVTT_CUE_IDENTIFIER_STRING);
private static final String WEBVTT_TIMESTAMP_STRING = "(\\d+:)?[0-5]\\d:[0-5]\\d\\.\\d{3}"; private static final String WEBVTT_TIMESTAMP_STRING = "(\\d+:)?[0-5]\\d:[0-5]\\d\\.\\d{3}";
private static final Pattern WEBVTT_TIMESTAMP = Pattern.compile(WEBVTT_TIMESTAMP_STRING); private static final Pattern WEBVTT_TIMESTAMP = Pattern.compile(WEBVTT_TIMESTAMP_STRING);
private static final Pattern MEDIA_TIMESTAMP_OFFSET = Pattern.compile(OFFSET + "\\d+"); private static final Pattern MEDIA_TIMESTAMP_OFFSET = Pattern.compile(OFFSET + "\\d+");
private static final Pattern MEDIA_TIMESTAMP = Pattern.compile("MPEGTS:\\d+"); private static final Pattern MEDIA_TIMESTAMP = Pattern.compile("MPEGTS:\\d+");
private static final String WEBVTT_CUE_TAG_STRING = "\\<.*?>";
private final boolean strictParsing; private final boolean strictParsing;
private final boolean filterTags;
public WebvttParser() { public WebvttParser() {
this(true); this(true, true);
} }
public WebvttParser(boolean strictParsing) { public WebvttParser(boolean strictParsing, boolean filterTags) {
this.strictParsing = strictParsing; this.strictParsing = strictParsing;
this.filterTags = filterTags;
} }
@Override @Override
@ -137,8 +145,15 @@ public class WebvttParser implements SubtitleParser {
// process the cues and text // process the cues and text
while ((line = webvttData.readLine()) != null) { while ((line = webvttData.readLine()) != null) {
// parse the cue identifier (if present) {
Matcher matcher = WEBVTT_CUE_IDENTIFIER.matcher(line);
if (matcher.find()) {
// ignore the identifier (we currently don't use it) and read the next line
line = webvttData.readLine();
}
// parse the cue timestamps // parse the cue timestamps
Matcher matcher = WEBVTT_TIMESTAMP.matcher(line); matcher = WEBVTT_TIMESTAMP.matcher(line);
long startTime; long startTime;
long endTime; long endTime;
String text = ""; String text = "";
@ -159,7 +174,7 @@ public class WebvttParser implements SubtitleParser {
// parse text // parse text
while (((line = webvttData.readLine()) != null) && (!line.isEmpty())) { while (((line = webvttData.readLine()) != null) && (!line.isEmpty())) {
text += line.trim() + "\n"; text += processCueText(line.trim()) + "\n";
} }
WebvttCue cue = new WebvttCue(startTime, endTime, text); WebvttCue cue = new WebvttCue(startTime, endTime, text);
@ -193,6 +208,19 @@ public class WebvttParser implements SubtitleParser {
return startTimeUs; return startTimeUs;
} }
protected String processCueText(String line) {
if (filterTags) {
line = line.replaceAll(WEBVTT_CUE_TAG_STRING, "");
line = line.replaceAll("&lt;", "<");
line = line.replaceAll("&gt;", ">");
line = line.replaceAll("&nbsp;", " ");
line = line.replaceAll("&amp;", "&");
return line;
} else {
return line;
}
}
protected void handleNoncompliantLine(String line) throws ParserException { protected void handleNoncompliantLine(String line) throws ParserException {
if (strictParsing) { if (strictParsing) {
throw new ParserException("Unexpected line: " + line); throw new ParserException("Unexpected line: " + line);