Using sed to get a specific data block from a webpage - China DOS Union Forum

For example: use SED to get 1: in "DOS Help Room", after "DOS Questions & Answers & Problem Discussion (Help Room) - China DOS Union Forum - China DOS Union's United DOS Forum", then 2: after "font-family: Tahoma, ", get the characters until the ending ";"

Because there are too many identical characters in a certain webpage, I used the source code of the CN-DOS troubleshooting section as an example

The string after 1: can't find any duplicate, so script conflicts are avoided

Not sure whether this explanation is clear enough.

That's still not clear enough. It would be best to post a piece of the source code and then mark the part that needs to be extracted.

The following is the webpage code
==========2008-02-23============ Giant  <a class="channel_keylink" href="http://www.txwb.com/Article/wyzy/wyzl/SFYX/200703/5605.html" target="_blank">Love Box</a>  <a class="channel_keylink" href="http://www.txwb.com/Article/wyzy/wyzl/GCYX/200703/5497.html" target="_blank">Loki</a>online Tears of Time  Paradise I  Wulin Waizhuan Super Runner Legend World MechWar Huaxia
<A class="" title="Article title: Employed

In SED syntax I can only go by the date (%date:~0,10%), because this is unique in the whole webpage code.

Goal: get all Chinese characters; after Huaxia, I don't want any Chinese after the code “”
Keep “Huaxia”

That makes it clear now, right

If it's still not clear, look below
The characters to extract: Giant
Love Box
Loki
Tears of Time
Paradise
Wulin Waizhuan
Super Runner
Legend World
MechWar
Huaxia

[ Last edited by jiulong on 2008-2-23 at 06:07 PM ]

No, this gets all Chinese characters. What I want is the Chinese characters from 2008- up to the first code.

Floor1　jiulong	Posted 2008-02-23 15:35
中级用户　Posts 183　Credits 396
For example: use SED to get 1: in "DOS Help Room", after "DOS Questions & Answers & Problem Discussion (Help Room) - China DOS Union Forum - China DOS Union's United DOS Forum", then 2: after "font-family: Tahoma, ", get the characters until the ending ";" Because there are too many identical characters in a certain webpage, I used the source code of the CN-DOS troubleshooting section as an example The string after 1: can't find any duplicate, so script conflicts are avoided Not sure whether this explanation is clear enough.

Floor2　lxmxn	Posted 2008-02-23 16:22
版主　Posts 4,938　Credits 11,386
That's still not clear enough. It would be best to post a piece of the source code and then mark the part that needs to be extracted.

Floor3　jiulong	Posted 2008-02-23 17:58
中级用户　Posts 183　Credits 396
The following is the webpage code <p><P><FONT color=#000000>==========2008-02-23============<BR>Giant  <a class="channel_keylink" href="http://www.txwb.com/Article/wyzy/wyzl/SFYX/200703/5605.html" target="_blank">Love Box</a>  <a class="channel_keylink" href="http://www.txwb.com/Article/wyzy/wyzl/GCYX/200703/5497.html" target="_blank">Loki</a>online<BR>Tears of Time  Paradise I  Wulin Waizhuan<BR>Super Runner Legend World MechWar<BR>Huaxia</FONT></P> <P><FONT color=#000000><A class="" title="Article title: Employed In SED syntax I can only go by the date (%date:~0,10%), because this is unique in the whole webpage code. Goal: get all Chinese characters; after Huaxia, I don't want any Chinese after the code “</FONT></P>” Keep “Huaxia” That makes it clear now, right If it's still not clear, look below The characters to extract: Giant Love Box Loki Tears of Time Paradise Wulin Waizhuan Super Runner Legend World MechWar Huaxia [ Last edited by jiulong on 2008-2-23 at 06:07 PM ]

Floor5　jiulong	Posted 2008-02-23 22:34
中级用户　Posts 183　Credits 396
No, this gets all Chinese characters. What I want is the Chinese characters from 2008- up to the first <FONT color=#000000> code.

Floor4　lxmxn	Posted 2008-02-23 19:06
版主　Posts 4,938　Credits 11,386