# Regex Problem **URL:** \ **Category:** Dev/API Corner\ **Tags:** plugin-dev\ **Created:** [December 16, 2010, 6:12pm UTC](https://forums.plex.tv/t/regex-problem/6940 "2010-12-16T18:12:12Z")\ **Posts on this page:** 7\ **Page:** 1

**Author:** ![mclog](https://avatars.discourse-cdn.com/v4/letter/m/b5a626/32.png) [@mclog](https://forums.plex.tv/u/mclog)\ **Post date:** [December 16, 2010, 6:12pm UTC](https://forums.plex.tv/t/regex-problem/6940/1 "2010-12-16T18:12:12Z")

It doesn't work Using regex to parse some HTML, importing the re library. No errors thrown, just doesn't return anything. I can run the regex and get the data parsed as I'd expect running the python in Eclipse, python 2.6 I think, any reason why it wouldn't work in Plex? Please no comments on the excellent quality of my regex ;) Here's the function I'm using, where SCHEDULE\_URL="http://www.liveonlinefooty.com/schedule.php": ```auto def DisplaySchedule(sender):

dir = MediaContainer(viewGroup="InfoList")

## Define regex patterns for parsing schedule info
channelPattern = '(.*?)'
rowAgainPattern = '

(.*?)

'
tablePattern = ' .* .*(.*):(.*?) .*(...) (..).. (...) .*(.*)
(.*)

'

## Open the schedule URL
schedule = HTTP.Request(SCHEDULE_URL)
# Get the local timezone
tz = time.timezone

# Parse each row in the table to get the match info
for rowIterate in re.findall(rowAgainPattern, schedule, re.DOTALL):
matchIterate = re.search(tablePattern,rowIterate)
matchhour = (matchIterate.group(1))
matchmins = (matchIterate.group(2))
matchwday = (matchIterate.group(3))
matchdate = (matchIterate.group(4))
matchmont = (matchIterate.group(5))
matchname = (matchIterate.group(6))
matchcomp = (matchIterate.group(7))
matchtime = time.ctime((time.mktime(time.strptime((matchmont + '/' + matchdate + '/10 ' + matchhour + ':' + matchmins + ':00'), "%b/%d/%y %H:%M:%S")))-tz)
# Parse each found channel to create a channel viewing item
for channelIterate in re.findall(channelPattern, rowIterate):
url = WATCHURL + channelIterate[1]
dir.Append(WebVideoItem(title=matchname, subtitle=matchtime + ' ' + channelIterate[1], url=url))

return dir ``` ---

**Author:** ![pierre1313](https://sea1.discourse-cdn.com/plex/user_avatar/forums.plex.tv/pierre1313/32/523_2.png) [@pierre1313](https://forums.plex.tv/u/pierre1313)\ **Post date:** [December 16, 2010, 7:44pm UTC](https://forums.plex.tv/t/regex-problem/6940/2 "2010-12-16T19:44:00Z")

> [@](#): > > Using regex to parse some HTML, importing the re library. > > No errors thrown, just doesn't return anything. I can run the regex and get the data parsed as I'd expect running the python in Eclipse, python 2.6 I think, any reason why it wouldn't work in Plex? Please no comments on the excellent quality of my regex ;) > > Here's the function I'm using, where SCHEDULE\_URL="http://www.liveonlinefooty.com/schedule.php": > > > ```auto > > def DisplaySchedule(sender):
>
> dir = MediaContainer(viewGroup="InfoList")
>
> ## Define regex patterns for parsing schedule info
> channelPattern = '(.*?)'
> rowAgainPattern = '

(.*?)

'
> tablePattern = ' > .* > .*(.*):(.*?) > .*(...) (..).. (...) > .*(.*)
(.*)

'
>
> ## Open the schedule URL
> schedule = HTTP.Request(SCHEDULE_URL)
> # Get the local timezone
> tz = time.timezone
>
> # Parse each row in the table to get the match info
> for rowIterate in re.findall(rowAgainPattern, schedule, re.DOTALL):
> matchIterate = re.search(tablePattern,rowIterate)
> matchhour = (matchIterate.group(1))
> matchmins = (matchIterate.group(2))
> matchwday = (matchIterate.group(3))
> matchdate = (matchIterate.group(4))
> matchmont = (matchIterate.group(5))
> matchname = (matchIterate.group(6))
> matchcomp = (matchIterate.group(7))
> matchtime = time.ctime((time.mktime(time.strptime((matchmont + '/' + matchdate + '/10 ' + matchhour + ':' + matchmins + ':00'), "%b/%d/%y %H:%M:%S")))-tz)
> # Parse each found channel to create a channel viewing item
> for channelIterate in re.findall(channelPattern, rowIterate):
> url = WATCHURL + channelIterate[1]
> dir.Append(WebVideoItem(title=matchname, subtitle=matchtime + ' ' + channelIterate[1], url=url))
>
> return dir > > ``` You have "import re" at the top of your file, right ? Also, are you opposed to the use of xpath and HTML.ElementFromURL ? That would make things simpler wouldn't it ? ---

**Author:** ![mclog](https://avatars.discourse-cdn.com/v4/letter/m/b5a626/32.png) [@mclog](https://forums.plex.tv/u/mclog)\ **Post date:** [December 16, 2010, 7:57pm UTC](https://forums.plex.tv/t/regex-problem/6940/3 "2010-12-16T19:57:47Z")

Yes, I have the import re, don’t known the xpath stuff at all, but I’m not strongly opposed to it, just that I know the regex I have is correct in python terms, just not plex plugin python ☹ ---

**Author:** ![sander1](https://avatars.discourse-cdn.com/v4/letter/s/4bbf92/32.png) [@sander1](https://forums.plex.tv/u/sander1)\ **Post date:** [December 16, 2010, 9:30pm UTC](https://forums.plex.tv/t/regex-problem/6940/4 "2010-12-16T21:30:27Z")

Without regexing, using Xpath, your code can look like this (untested): ```auto def DisplaySchedule(sender):

dir = MediaContainer(viewGroup="InfoList")

# Open the schedule URL
# When using PMS framework v1:
schedule = XML.ElementFromURL(SCHEDULE_URL, errors='ignore', isHTML=True)
# Or when using PMS framework v2:
#schedule = HTML.ElementFromURL(SCHEDULE_URL, errors='ignore')

for row in schedule.xpath('//div[@id="mycss0" or @id="mycss1"]/table'):
title = row.xpath('.//td[3]/span[@class="title"]')[0].text.strip() # Example: Juventus vs Man City
comp = row.xpath('.//td[3]/text()[1]')[0].strip() # Example: UEFA Europa League

time = row.xpath('.//td[1]')[0].text # Example: 18:00
date = row.xpath('.//td[2]')[0].text # Example: Thu 16th Dec
date = date.split(" ") # Example: ['Thu', '16th', 'Dec']

day = date[1][0:-2] # Get the day of the month and chop off last 2 characters (st/nd/rd/th)
month = time.strptime(date[2], "%b").tm_mon

# We can't get the year from the website, so determine the right year by checking if the scraped month is smaller than the current month
current_month = int(time.strftime("%m"))
year = int(time.strftime("%Y"))
if month < current_month:
# Month is in the next year
year = year + 1

timestamp = time.mktime(time.strptime(' '.join([str(day), str(month), str(year), time]), "%d %m %Y %H:%M")) - time.timezone
tuple = time.localtime(timestamp)
subtitle = time.strftime('%a %d %b, %Y; %H:%M', t)

for channel in row.xpath('.//a[contains(@href, "watchlive")]'):
url = WATCHURL + channel.get('href')
live_channel = channel.text
dir.Append(WebVideoItem(url, title=title + ' (' + comp + ')', subtitle=subtitle + ' ' + live_channel))

``` ---

**Author:** ![mclog](https://avatars.discourse-cdn.com/v4/letter/m/b5a626/32.png) [@mclog](https://forums.plex.tv/u/mclog)\ **Post date:** [December 17, 2010, 7:03pm UTC](https://forums.plex.tv/t/regex-problem/6940/5 "2010-12-17T19:03:10Z")

Thanks sander, you are legend! A couple of the xpaths don’t seem to work despite what xpath checker says, but pretty close anyway. I still have the question why my seemingly python correct regexes don’t work though 🙂 ---

**Author:** ![sander1](https://avatars.discourse-cdn.com/v4/letter/s/4bbf92/32.png) [@sander1](https://forums.plex.tv/u/sander1)\ **Post date:** [December 17, 2010, 10:33pm UTC](https://forums.plex.tv/t/regex-problem/6940/6 "2010-12-17T22:33:34Z")

> [@](#): > > I still have the question why my seemingly python correct regexes don't work though :) Not sure, but it could be a newline-related problem. ---

**Author:** ![system](https://global.discourse-cdn.com/plex/original/3X/2/a/2acb9765406f63293d357b4ec509ec39aa28f2ad.png) [@system](https://forums.plex.tv/u/system)\ **Post date:** [December 20, 2019, 8:55pm UTC](https://forums.plex.tv/t/regex-problem/6940/7 "2019-12-20T20:55:02Z")

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.