Skip to content

michael-bryson/Xml2Txt

 
 

Repository files navigation

Xml2Txt

The Xml2Txt application reads from a wikipedia dump file, and outputs a defined number of files at a defined minimum size.

The ./publish directory contains the executable zipped for simplicity.

The ./publish directory contains a small sample data.xml file of 10MB in size to get you moving.

publish the project, it should pick up the publish profile at Xml2Txt\Properties\PublishProfiles\FolderProfile3.pubxml

Download a dump file

.xml.bz2 compressed files can be downloaded from https://dumps.wikimedia.org/enwiki/

Example:/20210901/enwiki-20210901-pages-articles-multistream.xml.bz2

This is an 18GB compressed file containing some 25,000,000 wikipedia pages.

Use a utility such as 7-zip to extract this to enwiki-20210901-pages-articles-multistream.xml,

Rename the extracted file to something simpler, like data.xml.

The extracted file is now some 90GB.

If you want to inspect the structure use something like http://www.ximple.cz to split the file into manageable chunks.

Copy the ../publish/Xml2Txt.exe to the same directory as the .xml file.

Click on Xml2Txt.exe and you will be prompted for:

  • The number of files to create.
  • The minimum size of each file in MB.
  • The data file to use (maybe data.xml).
  • A file prepend/directory name to use when generating files.

The application will read each page from the xml, when it has read content greater than the minimum file size it will create a file in the defined directory.

Example: 5 files, min size 10 MB, using data.xml as input, writing to directory "BIG" will produce:

  • BIG
    • BIG_1.txt
    • BIG_2.txt
    • BIG_3.txt
    • BIG_4.txt
    • BIG_5.txt

Image showing operation

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C# 100.0%