I'm doing an app which can parse XML files generated from different programs. I need to check the first line of the file to detect which program has generated the xml and call the correct method to parse it.
i.e one of the file generated starts with this line:
<PROFILE XYZ="1">
another program generates the file starting with this line:
<AppXYZ DBVersion="1.2.3.4">
I need to check this line.
Any help is appreciated.
Thanks,
Max
It's XML, so just parse it as XML. Check the first element and decide from there. SAX or DOM, either works.
Related
We are generating xlsx files using a perl script. Files usually contains thousands of records. This makes spotting errors a very difficult operation.
This process was working since years without problems.
This week we got a request to check a file which contains errors. While opening Excel prompted that the file contains errors and asked whether we want to repair them.
In fact we do not want to recover the data but want to know which part of the file is corrupt. The error should be coming from corrupt data and we are interested to identify these data.
the log message shows the following:
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<recoveryLog xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
<logFileName>error068200_01.xml</logFileName> </br>
<summary>Errors were detected in file 'D:\Temp\20161020\file_name.xlsx'</summary>
<repairedRecords summary="Following is a list of repairs:"><repairedRecord>Repaired Records: Cell information from /xl/worksheets/sheet1.xml part</repairedRecord>
</repairedRecords>
</recoveryLog>
The error should come from corrupt data. Is there any tool/method which helps to spot this corrupt data?
I tried renaming it a zip file, extracting it and opening it via an XML editor but was not able to find any errors in XML file.
We also checked that the different XML file structures are fine.
Thank you and best regards
As expected, the problem was coming from text cells containing numbers having an E in the middle.I used the following steps to identify the erronous cells.
1. Wrote small Java class to read the file. The class was checking the cell type and displaying the value afterwards.The java program generated an Exception at some line "Cannot get a numeric value from a text cell" even If I was correctly checking the cell type before displaying the content.
2. I checked the opened Excel file at that line and found that the cell contains only 'inf'.
3. I opened the file using open office and looked at the same cells. They contain 0.
4. I debugged the program generating the data and found out that these cells contain data like '914E5514'. Seems that E which was interpreted by Excel as an exponent.We changed the program to use the format '#' for that cell and this solved the issue.
Thank you.
Thank you very much, you helped me a lot by saying that 1 particular content item may be the root problem.
My corrupted content was https://www.example.com XYZ ... ASDAS
Solution: www.example.com XYZ ... ASDAS
This is something which cannot be handled by excel. Would be nice to have a list of thing which do not work
I have a file that I use as a template, which is copied over to each one of my customer folders. I manually go into the file and edit one line EarNotch=5/5 to EarNotch=13/6. Is there a way I can add it to my current PowerShell script that copies over the file that would edit the file with the new unique ID that I require for this customer?
The file type I am working with is a .xml file. The EarNotch is always on Line 2. Let me know what else you might need to know to help me figure this out.
The simplest way is to load the xml file using Get-Content, replace the string, then cast as xml so you can consume properly. This should answer your question.
$fileContents = Get-Content -Path C:\path\to\file.xml
[xml]$xmlFile = $fileContents.Replace('EarNotch="5/5"','EarNotch="13/6"')
For anything more complex, you might want to use XQuery, or other xml parsing.
Without knowing the structure of your xml file, the people on stack overflow can't help you. It's always best to post as much info as possible.
Actually I know that there is Test::Valgrind::Parser::XML perl module. But I have no idea how to use it: If anyone can provide documentation it would be great.
The valgrind docs show that valgrind accepts a --xml=yes tag to output messages as XML. The format of the XML is specified in the docs/internals/xml-output-protocol4.txt inside the source code repository.
With that, you can use any XML parser and do whatever you want with the data.
I am trying to write a script that will allow me to download numerous (1000s) of data files from a data server (e.g, http://hydro1.sci.gsfc.nasa.gov/thredds/catalog/GLDAS_NOAH10SUBP_3H/2011/345/). Unfortunately, the names of the files in each directory are not formatted in a similar way (the time that they were created were appended to the end of the file name). I need to be able to specify the file name to subset the data (I have a special tool for these data types) and download it. I cannot find a function in matlab that will extract the file names.
I have looked at URLREAD, but it downloads everything including html code.
Thanks for your help!
You can easily parse the link.
x=urlread(url)
links=regexp(x,'<a href=''([^>]+)''>','tokens')
Reads every link, you have to filter all unwanted links.
For example this gets all grb files:
a=regexp(x,'<a href=''([^>]+.grb)''>','tokens')
I am engaged in preparing an application regarding reading the .epub files in iPhone. Where can I get the reference for sample applications for unzipping and parsing the files? Can anyone guide me with a best link? Thank you in advance.
An .epub file is just a .zip file. It contains a few directory files in XML format and the actual book content is usually XHTML. You can use Objective-Zip to unzip the .epub file and then use NSXMLParser to parse the XML files.
More info: Epub Format Construction Guide
On top of Ole's answer (that's a pretty good how-to guide), it's definitely worth reading the specification for the Open Container Format (OCF) - sorry it's a word file. It's the formal specification for the for zip structure used.
In brief you parse the file by
Checking it's plausibly valid by looking for the text 'mimetype' starting at byte 30 and the text 'application/epub+zip' starting at byte 38.
Extracting the file META-INF/container.xml from the zip
Parsing that file and extracting the value of the full-path attribute of the first rootfile element in it.
Load the referenced file (the full-path attribute is a URL relative to the root of zip file)
Parse that file. It contains all the metadata required to reference all the other content (mostly XHTML/CSS/images). Particularly you want to read the contents of the spine element which will list all content files in reading order.
If you want to do it right, you should probably also handle DTBook content as well.
If you want to do this right, you need to read and understand the Open Packaging Format (OPF) and Open Publication Structure (OPS) specifications as well.