Replacing specific text from document A to B

[Václav Trunec]

Hi there, I was wondering if anyone can help me do this:

I have two documents opened side by side (as shown below):

I am trying to get the already-translated text from document A (on the left) to the respective lines in document B. I cannot just copy-paste this, because I know there might be some lines that can have different “keys” and are on different lines.
The thing is (as you can see on the screenshot) I cannot just select lines from document A and paste them into document B, because they are on the different lines.

My question is: Is there a way to replace text in the document B if it matches the “key” in the document A? Such as (as for the screenshot):

• I have a key called “searchInputHintOnTable” in the document A. If there is the same key in the document B, use the text ‘Search by alert name’ (from document A) to replace the ‘Hledat podle názvu’ (in document B).

I hope I explained it well enough, but I will be more than happy to specify this as much as possible.

Thank you.

[Alan Kilborn]

You really should follow the instructions HERE for posting such a question.

[guy038]

Hello, @Václav-Trunec and All,

I’ve got a solution to your problem which is a bit long, but without any major difficulty !

It will use :

• 5 regex search/replacements

• 2 sort operations

• The use of the column editor to number the lines

---

I assume that your initial Document A contains only lines like :

Key: 'English strings',

and that your initial Document B contains only lines like :

Key: 'Czech strings',

A last advice : Better to read carefully all this post first, before beginning any operation !!

---

Regarding the 5 regex S/R, you’ll follow these steps :

• Open the Replace dialog ( Ctrl + H )

• Type in the SEARCH regex in the Find what: field

• Type in the REPLACE regex in the replace with: field

• Untick all options

• Tick the Wrap around option

• Select the Regular expression search mode

• Click preferably on the Replace All button ( Do not use the Replace button in the fourth regex S/R )

---

Let’s go :

• Copy/paste the Document A contents in a new Document C

• Use this first regex S/R, in Document C :

  • SEARCH ^(\w+):

  • REPLACE \1 CCC:

=> Contents of Document C should look as below :

...
...
searchInputHintOnTable CCC: 'Search by alert name',
searchInputHint CCC: 'Find the alert belonging to a specific alert by its name.',
...
...

• Copy/paste the Document B contents in a new Document D

• Use this second regex S/R, in Document D :

  • SEARCH ^(\w+):

  • REPLACE \1 DDD:

=> Contents of Document D should look as below :

...
...
searchInputHint DDD: 'Najděte upozornění patřící konkrétnímu alertu podle jeho názvu',
searchInputHintOnTable DDD: 'Hledat podle nazvu alertu',
...
...

• Now, try to find out the longest line of Document D. Its length will be set to the Lm variable

• Move to the first line of Document D. Its length will be set to the L1 variable

• Determine the value X = Lm - L1 + 5

• Move at the end of the first line of Document D

• Add X space chars

• Open the Column Editor ( Alt + C )

  • Choose the Number to Insert option

  • Type in the value 1 in the Initial number : field

  • Type in the value 1 in the Increase by : field

  • Type in the value 1 in the Repeat : field

  • Tick the Leading zeros option ( IMPORTANT )

  • Verify that the Dec option is selected

  • Click on the OK button

=> A numbering should appear at the end of each line of Document D

...
...
searchInputHint DDD: 'Najděte upozornění patřící konkrétnímu alertu podle jeho názvu',     xxxx
searchInputHintOnTable DDD: 'Hledat podle nazvu alertu',                                   yyyy
...
...

• Merge document C and Document D in a new Document E

=> Contents of Document E should look as below :

...
...

searchInputHintOnTable CCC: 'Search by alert name',
searchInputHint CCC: 'Find the alert belonging to a specific alert by its name.',
...
...
searchInputHint DDD: 'Najděte upozornění patřící konkrétnímu alertu podle jeho názvu',     xxxx
searchInputHintOnTable DDD: 'Hledat podle nazvu alertu',                                   yyyy
...
...

• Now, in Document E :

  • Suppress any selection

  • Use the Edit > Line Operations > Sort Lines Lexicographically Ascending option

=> The contents of Document E become :

...
...
searchInputHint CCC: 'Find the alert belonging to a specific alert by its name.',
searchInputHint DDD: 'Najděte upozornění patřící konkrétnímu alertu podle jeho názvu',     xxxx
....
....
searchInputHintOnTable CCC: 'Search by alert name',
searchInputHintOnTable DDD: 'Hledat podle nazvu alertu',                                   yyyy
....
....

• Use this third regex S/R, in Document E :

  • SEARCH (?-is)^(.+) CCC: (.+)\R\1 DDD: \K.+,

  • REPLACE \2

=> The contents of Document E are now changed into :

...
...
searchInputHint CCC: 'Find the alert belonging to a specific alert by its name.',
searchInputHint DDD: 'Find the alert belonging to a specific alert by its name.',     xxxx
...
...
searchInputHintOnTable CCC: 'Search by alert name',
searchInputHintOnTable DDD: 'Search by alert name',                                   yyyy
...
...

• Use this fourth regex S/R, in Document E :

  • SEARCH (?-is)^.+ CCC: .+\R|^(.+ DDD: .+?)\x20+(\d+)$

  • REPLACE (?1\2 \1:)

=>

• All the lines containing the CCC: string, surrounded by space chars, should be deleted :

• Numbering, in lines containing the DDD: string, surrounded by space chars, should have moved from the end to the beginning of each line

• Remember that xxxx and yyyy are numbers !

...
...
xxxx    searchInputHint DDD: 'Find the alert belonging to a specific alert by its name.',
...
...

yyyy    searchInputHintOnTable DDD: 'Search by alert name',
...
...

• Again, in Document E :

  • Suppress any selection

  • Use the Edit > Line Operations > Sort Lines Lexicographically Ascending option

=> This time, all the lines should be sorted, according to their initial location in Document B and Document D

...
...
xxxx    searchInputHint DDD: 'Find the alert belonging to a specific alert by its name.',
yyyy    searchInputHintOnTable DDD: 'Search by alert name',
...
...

• Use this fifth and final regex S/R, in Document E :

  • SEARCH (?-i)^\d+\x20{4}|\x20DDD(?=:\x20)

  • REPLACE Leave EMPTY

Your final Document E should look like below :

...
...
searchInputHint: 'Find the alert belonging to a specific alert by its name.',
searchInputHintOnTable: 'Search by alert name',
...
...

• The order of lines, in Document E, should be identical to the Document B order

• The Czech version of all lines, in Document B and Document E, should have been replaced with their English equivalents !

Best Regards

guy038

[Václav Trunec]

First of all, thank you very much for the thorough explanation, it is absolutely wonderful.

However, I am stuck at this thing in Document E:

• Use the Edit > Line Operations > Sort Lines Lexicographically Ascending option

If I do this, the document will not be usable for our developers, since it also moves the { and } characters, determining the end of a translation key. Is there a way to skip this?

Thank you again.

[PeterJones]

@václav-trunec said in Replacing specific text from document A to B:

If I do this, the document will not be usable for our developers, since it also moves the { and } characters, determining the end of a translation key. Is there a way to skip this?

If you follow Guy’s full instructions, you will see that he added a “key” (the extra text) which will allow his two different sorts to undo each other: he sorts on that “key”, and then in the end, it re-sorts back to the original order. As Guy said, “The order of lines, in Document E, should be identical to the Document B order” after his last step.

-----

Alternatively, I recently published a script in a similar discussion, which will allow you set up pairs of English/OtherLanguage translation strings (or, in your case, OtherLanguage-to-English). If you follow the instructions in that post, but use the original/Czech as the left side of the translation dictionary and the translated/English as the right side, it should work.

So, instead of using the example from that discussion:

translation = {
    r'"JJIS number:"':              r'"د JJIS شمیره:"',
    r'"Choose one"':                r'"یو يې غوره کړئ"',
    r'"Social Security number:"':   r'"د ټولنیز مصؤنیت لمبر:"',
    r'"State ID \(SID\):"':         r'"د ایالت ID \(SID\):"'
}

… you would use something like …

translation = {
    r'Najděte upozornění patřící konkrétnímu alertu podle jeho názvu':              r'Find the alert belonging to a specific alert by its name.',
    r'Hledat podle nazvu alertu':         r'Search by alert name'
}

(remembering to end every pair except the last one with a comma