• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 167
  • Last Modified:

How to strip down cURL return to just the asked for field data?

This cURL script works fine for accessing my API:

curl -X GET -H "ACCEPT: application/xml" -H "X-AUTH-TOKEN: cee8d8a1afd5921dc1efe1817"  "https://api.niftywebsite.com/?opts=APNMB&apps=13629198"  -o out.txt

APNMB is a field in the database.

What I get back is:

<?xml version="1.0" encoding="utf-8"?>
<response>
      <entity>
            <APNMB>13629198</APNMB>
      </entity>
</response>

All I want back is the app number, which is 13629198

Other than writing to a temp file and stripping it out, is there a way directly in cURL to just get the app number returned?
0
Mike Caldwell
Asked:
Mike Caldwell
  • 7
  • 4
  • 3
3 Solutions
 
nociSoftware EngineerCommented:
No,

curl is meant to do just the transfer parts. I is not meant to handle content.
otoh, by running:

curl <yourquery> | parse-script

where parse script can be:
---8<---
#!/bin/bash
grep 'APNMB' | sed -e 's/^[^>]*>//' -e 's/<.*$//'
---8<---

or just:
curl <yourquery> |grep 'APNMB' | sed -e 's/^[^>]*>//' -e 's/<.*$//'
0
 
shalomcCommented:
No, there is no way for curl to do it directly.
You don't have to write to a temp file, though. You can pipe it to an XML processing script/program/utility as input.
Just don't use the -o switch.

curl -X GET -H "ACCEPT: application/xml" -H "X-AUTH-TOKEN: cee8d8a1afd5921dc1efe1817"  "https://api.niftywebsite.com/?opts=APNMB&apps=13629198"      | some-xml-processor
0
 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
noci, I added |grep 'APNMB' | sed -e 's/^[^>]*>//' -e 's/<.*$//'  after my full query, and I got an error message of "grep is not an internal or external command"
0
[Webinar] Cloud and Mobile-First Strategy

Maybe you’ve fully adopted the cloud since the beginning. Or maybe you started with on-prem resources but are pursuing a “cloud and mobile first” strategy. Getting to that end state has its challenges. Discover how to build out a 100% cloud and mobile IT strategy in this webinar.

 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
OK, new learning!  I did not realize that "|" meant new line.  So I redid it, and I still get that error message.
0
 
nociSoftware EngineerCommented:
| is NOT newline, but send output from command before | to input of command after |.

It does mean end of a command part.  And all input needs to be in ONE line.
0
 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
That is how I have it, and I still get the whole structure, not just the part I want,.  I think I'm hoping for something that cURL won't do.
0
 
nociSoftware EngineerCommented:
curl cannot parse xml...

curl -X GET -H "ACCEPT: application/xml" -H "X-AUTH-TOKEN: cee8d8a1afd5921dc1efe1817"  "https://api.niftywebsite.com/?opts=APNMB&apps=13629198" | grep 'APNMB' | sed -e 's/^[^>]*>//' -e 's/<.*$//'

All on one line should only produce the number:
13629198

note: the " en ' characters ARE important.

if not then please produce wat IS shown as an answer including error messages etc.
0
 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
Got an error message that grep is not a recognized internal or external program.  Windows cURL seems to require double quotes, so I copied and pasted what you posted, then tried again with double quotes like this:


curl -X GET -H "ACCEPT: application/xml" -H "X-AUTH-TOKEN: ceed8d98159b218dc1efe1817" "https://api.niftysite.com/v3/appSearch/?opts=APBNMB&apps=13629198" | grep "APNMB" | sed -e "s/^[^>]*>//" -e "s/<.*$//" -o EE.txt
0
 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
To be clear: nothing shown since it did not run.
0
 
shalomcCommented:
grep and sed are unix/linux commands, so if you run it on windows you get nothing.

Here is a XSLT transformation to retrieve your value:
<?xml version="1.0"?>   
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">  
<xsl:template match="APNMB">       
<xsl:value-of select="."/>   
</xsl:template>                    
</xsl:stylesheet>                

Open in new window


All you need now is an XSLT processor.
Try this https://sourceforge.net/projects/saxon/files/Saxon-HE/

I will post shortly more detailed instructions..
0
 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
Please do; I did not see anything in the download that hinted at installation.
0
 
nociSoftware EngineerCommented:
old_mike your are correct, cURL is a transfer program so you can access remote systems from a command line or a script... so can't handle content... like html, text, ....  just transfer methods like
HTTP, FTP, ... so how should it hint about content. You can build complete scripts (except for javascript handling) in a shell language, (or php, or perl) to simulate clicking through websites by filling out webforms, GET-ting info, POST-ing info etc. all while managing the cookies & other HTTP headers.
The tool was developed on Unix and ported to various platforms. On Unix (Linux too) there are a lot of programs with a specific defined task. so cURL is a transfer tool, grep is a search tool (like Find on windows, a tad more powerful though), and sed is a kind of editor.

Also getting the info using an XSLT transform will be more exact. The grep / sed method works for a known case with a simple value like your problem.

BTW. For unix tooling under windows checkout cygwin. ( http://cygwin.org/ )
curl & bash can be part of that.

BTW2, it helps to also describe the platform in the question.
0
 
Mike CaldwellDirector of Business Development, AnaquaAuthor Commented:
Yeah, I see that I should have mentioned Windows up front.  Noci, this project is getting way beyond me from a time standpoint.  My partner is my son, and he is expert in all this, just doing something else so I thought I would make some very simple reports from the tool.  By the time I get done, he'll have it done, so I think I'll thank everyone and stop here.  I may still try to access the API with a VBScript, which I have done a lot of.
0
 
shalomcCommented:
Installing Saxon:
https://sourceforge.net/projects/saxon/files/Saxon-HE/9.6/
Get the "setup" exe file. It will create a folder in "program files". Add the bin folder you your path.

Running XSLT from the command line:
http://www.saxonica.com/html/documentation/using-xsl/commandline.html

Save the XSLT into a file, for example old_mike.xslt and run
Transform -s:xmlfile.xml  -xsl:old_mike.xslt -strip:all

Add the -o:outputfile switch if you need it in a file.

Here is an updated XSLT file.  
<?xml version="1.0"?>
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="2.0">  
<xsl:output method="text"/>
<xsl:template match="APNMB">       
<xsl:value-of select="."/>   
</xsl:template>                    
</xsl:stylesheet>                

Open in new window

0

Featured Post

How to Use the Help Bell

Need to boost the visibility of your question for solutions? Use the Experts Exchange Help Bell to confirm priority levels and contact subject-matter experts for question attention.  Check out this how-to article for more information.

  • 7
  • 4
  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now