Find parent tag's id if that parent tag has specific tags inside it

dinkumoil

Please note: The following is only true if you accidentally wrote <revst> instead of <revst/> or <revst> ... </revst>. If your document only contains <revst>, it is not a valid XML document because it lacks the related closing tag and the method I’m explaining below doesn’t work.

A more reliable and much more hassle-free method to search for things in XML documents is XPath, a query language that was explicitely designed for that use case.

To be able to use XPath you need to install the XML Tools plugin, available via Npp’s build-in Plugins Admin. After installing it, navigate to (menu) Plugins -> XML Tools -> Evaluate XPath expression. In the dialog box popping up, enter //task[descendant::revst]/@key into the upper input field. The plugin will generate the result in the lower ListView.

You can copy the result to your clipboard by clicking the according button. After pasting it to an empty document, you only have to do some column selection to remove the content of the ListView’s first two columns, which are useless for you but have been copied to the clipboard as well.

ganesan govindarajan

Hi @guy038,

Thanks much for your help its working fine!!.

Somewhat i didn’t realized that format of task may differ file to file as key attribute position is may change task to task as like below,

<task breaknbr="00" chapnbr="71" chg="U" func="000" key="TASK-710000000800" pgblknbr="301" revdate="19950301" sectnbr="00" seq="800" subjnbr="00">

Key attribute comes at 4th or 5th position or dynamically changed file to file. Due to this above regex cannot find in some of the sgm files.

Can you please help me out on this?

Thanks again.

ganesan govindarajan

Hi @dinkumoil ,

Thanks for asking!!

Actually we don’t worry about the closing tag as we need to find only opening tag. Also this is a sgm file and closing tag is different “<revend>”.

thanks.

guy038

Hello @ganesan-govindarajan and All,

Sorry, but I do not understand :-(

If I use, for instance, this text :

<task breaknbr="00" chapnbr="71" chg="U" func="000" key="TASK-710000000800" pgblknbr="301" revdate="19950301" sectnbr="00" seq="800" subjnbr="00">
<title>AAA</title>
<para>BBAACC</para>
<para><revst>BBAACC</para>
</task>

The regex does mark the string TASK-710000000000 ? No difference since previously ! I probably miss some details.

BR

guy038

ganesan govindarajan

Hi @guy038 ,

When i tried to use the regex it is only selected the first line of file. Thats the reason, i thought the issue occurs because of task format.

I don’t know why it is selected only first line of file.

ganesan govindarajan

Hi @guy038 ,

Do we need to main the same format of text as per above example underneath the “task”? because lot of contents and tags find my new file after the task element.

guy038

Hi, @ganesan-govindarajan and All,

I’ll try to briefly explain how this regex works and you should be able to verify if it meets your needs !

So, the regex (?-si)^\h*<task .+"\KTASK.+?(?=(?s:"((?!</task>).)+?)<revst>), used with the Mark dialog, means :

From the beginning of a line, it first finds optional leading blank characters, followed with the string <task, in lower case, followed with a space char, followed by anything till a double quote ( The one right before the string TASK ! )
Then the \K regex syntax resets the present search. Thus, it just looks for the string TASK, in upper case, followed with any non-null range of chars, till the nearest double quote char ( The one right after the string TASK-xxxxxxxxxx )
But ONLY IF two additional conditions exist :
- It must be followed, further on, with a <revst> tag, first
- Then, it must be followed, further on, with the ending tag </task>

This implies that, IF the </task> string is found before a <revst> tag, the regex will not match ! )

Globally, it can find any TASKxx....xxx string, if it is followed, downwards, with a <revst tag then, downwards, with an ending </task> tag ! So :

You may have many lines, or none, between the line <task.........> and the <revst> tag
You may have many lines, or none, after the <revst> tag till the ending tag </task>

For instance, even the short line, below :

<task key="TASK:AB12345"><para><revst>BBAACC</para></task>

Or this minimum one :

<task key="TASK:AB12345"><revst></task>

would mark the string TASK:AB12345 !

Of course, I assume that :

The part <task ................> begins a line, after possible blank characters, only
The part key="TASKxxxxx" is located on a single line

If other cases may occur, just tell me !

Best Regards,

guy038

ganesan govindarajan

Hi @guy038 ,

Your explanation is so well!! thanks much for that!!

This implies that, IF the </task> string is found before a <revst> tag, the regex will not match ! )

Yes <revst> tag may appear before or after the <task—></task> element inside of file as other elements are also have the same <revst> tags. Is this the reason this regex is not supported?

ganesan govindarajan

Hi @guy038

May be my tasks starts and ends with as below,

guy038

Hi, @ganesan-govindarajan,

Actually, if a <revst> tag is found outside a <task.....>........</task> section OR if <task ........>........</task> sections do not contain, at least, one <revst> tag, like below :

...
<task .......>
....
....
....
</task>
...
...
...<revst>...
...
...
<task .......>
....
....
....
</task>
...

The regex will not match anything ! I suppose that this behavior is the one which is expected !?

Now, I need additional text to get a general idea of what you need. Could you provide a real complete example, as a text, in reverse video ?

Could you indicate the zones to mark and the different mandatory conditions to respect ?

Thanks,

BR

guy038

ganesan govindarajan

Hi @guy038 ,

I really sorry for the late response.

I am just trying to share the real time data sample here to understand clearly.

<em ..............>
.......................
<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<effect effrg="710/ALL" efftext="710/ALL"></effect>
<title>sample text</title>
<subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<title>sample text</title>
<para><revst>sample para<revend></para>.....
</subtask>
</task>
<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200001" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<effect effrg="710/ALL" efftext="710/ALL"></effect>
<title>sample text</title>
<subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200003" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<title>sample text</title>
<para>sample para</para>.....
</subtask>
</task>

.
.
.
.
.
.
.
.
</em>

As per the above example, the xml file may contain one or more task continuously. Since, I need to find <revst> in between the <task>…</task> and to get respective task “key” value. If any task does not contain <revst> tag we dont need that task’s key value.

In the above example, the expected output is: TASK-712101200000 as <revst> found in the first task.

Your previous regex works correctly but that works only for the fist instance of task and not finding rest of the instances.

I hope you understand . Kindly let me know if any data required.

Thanks for your patience!
Ganesan G

guy038

Hello, @ganesan-govindarajan,

Indeed, it’s been a very long time since your last post, in this other topic :

https://community.notepad-plus-plus.org/post/80495

Which seems related to your same problem !

So, from your example, I understood that you want the TASK-712101200000 value because the curent <task.........</task> section contains the <revst> tag !

Now, this section contains the TASK-712101200000 value, twice. So :

Do you want the regex to match the two values TASK-############ ?
Do you want the regex to match the first TASK-############ value, only ?

And, if the regex must match all the instances of the block, may this case happen :

<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<effect effrg="710/ALL" efftext="710/ALL"></effect>
<para><revst>sample para<revend></para>.....
<title>sample text</title>
<subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<title>sample text</title>
</subtask>
</task>

Where the <revst> tag comes after the first TASK-############, but before the second TASK-############ ?

Best Regards,

guy038

ganesan govindarajan

Hi @guy038 ,

Thanks for the response after a long time. Yes its been a very old post that have asked for.

I need your help on this!!

Actually each TASK, SUBTASK’s key values are unique and does not repeat or duplicate anywhere at any case. Even though all type of element’s key values are unique. That was my typo sorry for that.

In my case, I need only the TASK’s key value irrespective of other key values inside the <task>…</task> if “<revst>” tag(s) is exists.

Also the existing one that i have asked and routed by you (post 80495) is not exactly match with this requirement, Thats why i have raising here.

Thanks again!!
Ganesan. G

guy038

Hi, @ganesan-govindarajan,

Ah…, OK ! So, I duplicated your recent example, 3 times, giving the text below, with some leading indentations :

<em ..............>
    .
    .
    <task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
        <effect effrg="710/ALL" efftext="710/ALL"></effect>
        <title>sample text</title>
        <subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200002" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
            <title>sample text</title>
            <para><revst>sample para<revend></para>.....
        </subtask>
    </task>
    
    <task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200001" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
        <effect effrg="710/ALL" efftext="710/ALL"></effect>
        <title>sample text</title>
        <subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200003" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
            <title>sample text</title>
            <para>sample para</para>.....
        </subtask>
    </task>
    

    <task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101300000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
        <effect effrg="710/ALL" efftext="710/ALL"></effect>
        <title>sample text</title>
        <subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101300002" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
            <title>sample text</title>
            <para><revst>sample para<revend></para>.....
        </subtask>
    </task>
    
    <task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101300001" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
        <effect effrg="710/ALL" efftext="710/ALL"></effect>
        <title>sample text</title>
        <subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101300003" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
            <title>sample text</title>
            <para>sample para</para>.....
        </subtask>
    </task>

    
    <task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101400001" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
        <effect effrg="710/ALL" efftext="710/ALL"></effect>
        <title>sample text</title>
        <subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101400003" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
            <title>sample text</title>
            <para>sample para</para>.....
        </subtask>
    </task>

    <task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101400000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
        <effect effrg="710/ALL" efftext="710/ALL"></effect>
        <title>sample text</title>
        <subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101400002" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
            <title>sample text</title>
            <para><revst>sample para<revend></para>.....
        </subtask>
    </task>
    .
    .
</em>

To get a fair example, I also modified, on purpose, the numbers after TASK- string, in order that they are all unique :

TASK-712101200000
TASK-712101200002
TASK-712101200001
TASK-712101200003
TASK-712101300000
TASK-712101300002
TASK-712101300001
TASK-712101300003
TASK-712101400001
TASK-712101400003
TASK-712101400000
TASK-712101400002

I also swapped the two blocks of the last section, so :

The <task.... </task> block not containing the <revst> tag, comes first
The <task.... </task> block containing the <revst> tag, comes in second

And finally, in the last <task.... </task> block, I move up the line <para><revst>sample para<revend></para>..... right after the line :

<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101400000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">

And I still confirm that my regex, provided in my last post and reported below, does finds :

The first occurrence of the key value TASK-############, if the present <task.........</task> block does contain a <revst> tag
Do not find anything if if the present <task.........</task> block does not contain the <revst> tag

SEARCH / MARK : (?-si)^\h*<task .+"\KTASK.+?(?=(?s:"((?!</task>).)+?)<revst>)

To be convinced :

Paste the above text, in a new tab ( Ctrl + N )
Open the Mark dialog ( Ctrl + M )
Un-tick all box options
Tick the Wrap around option
Type in the regex (?-si)^\h*<task .+"\KTASK.+?(?=(?s:"((?!</task>).)+?)<revst>) in the Find what: zone
Select the Regular expression search mode
Click on the Mark All button

=> You should get the message : Mark: 3 matches in entire file

Click on the Copy Marked Text button
Paste the results in the new tab

=> You should get the additional text, below :

TASK-712101200000
TASK-712101300000
TASK-712101400000

Which corresponds to the first 3 occurrences, found in sections which do contain a <revst> tag !

Tell me if something seems unclear or if I still misunderstood the whole story !

BR

guy038

ganesan govindarajan

Hi @guy038

Thank you so much !! for the prompt responses.

In the above,

I also swapped the two blocks of the last section, so :

The <task.... </task> block not containing the <revst> tag, comes first

The <task.... </task> block containing the <revst> tag, comes in second

And finally, in the last <task.... </task> block, I move up the line <para><revst>sample para<revend></para>..... right after the line :

<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101400000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">

This shuffles may not possible the reason is tons of task will be there in the single file and unable to find it.

Kindly clarify on this? maybe i had overlooked on this.

thanks
Ganesan.G

guy038

Hello, @ganesan-govindarajan,

Well, when I decided to modify your example, by copying it three times, then using only unique key-values, swapping the cases which does and does not match and finally moving up the line containing the <revst> tag, it was just in order to show you that my regex did work with a more general text. So just a proof, in some way !

Now, I don’t want you to be upset by my modifications. So, I can just use your original text which is :

<em ..............>
.......................
<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<effect effrg="710/ALL" efftext="710/ALL"></effect>
<title>sample text</title>
<subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<title>sample text</title>
<para><revst>sample para<revend></para>.....
</subtask>
</task>
<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200001" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<effect effrg="710/ALL" efftext="710/ALL"></effect>
<title>sample text</title>
<subtask breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200003" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">
<title>sample text</title>
<para>sample para</para>.....
</subtask>
</task>

.
.
.
.
.
.
.
.
</em>

And using my regex against the abvove text, it would find 1 occurrence only ( TASK-712101200000 ) on the first line :

<task breaknbr="00" chapnbr="71" chg="U" func="200" key="TASK-712101200000" pgblknbr="801" revdate="20191115" sectnbr="21" seq="000" subjnbr="01">

Now, your example cannot be qualified as a real example. I need a bunch of text of, let’s say, 100 to 500 lines in order to safely test my regex in real conditions !

Best regards,

guy038

ganesan govindarajan

Hi @guy038

Thanks for checking!

I will provide the real time example with more content later.

As of now i got an idea based on your existing suggestions to get those IDs as following.

Mark all <task …>…</task>.
Mark all keywords to search <revst>.
Now “Copy marked texts”
Paste into new file and Mark all <task…>…</task> with <revst>
Now again “copy marked text”
Paste into new file and remove <revst> and get those <task…>…</task> key value now.

Please suggest if you have better way!!

thanks
Ganesan G

guy038

Hello, @ganesan-govindarajan,

Yes, I understand your point of view ! So, just follow this road map :

Open your file in Notepad++
Open the Mark dialog ( Ctrl + M )
Untick all box options
Enter the regex (?s)<task (?:(?!</task).)+?<revst>.+?</task> in the Find what: zone
Tick the Wrap around option
Select the Regular expression search mode
Click on the Mark All button
Click on the Copy Marked Text button
Open a new file ( Ctrl + N )
Paste the clipboard contents ( Ctrl + V ) in this new file
Re-open, if necessary the Mark dialog, with this new text as current file
Enter the regex <task .+?\Kkey=".+?" in the Find what: zone
Keep the other options unchanged
Re-click on the Mark All button
Re-click on the Copy Marked Text button
Open a second new file ( Ctrl + N )
Paste the clipboard contents ( Ctrl + V ) in this last new file

Here you are : you should get all the key values !!

Best Regards,

guy038

ThosRTanner

See also the faq about not using regexs to parse xml documents.