Find and Replace from Lists

[moldy]

Hi Everyone

I am new to regex so don't know if what I am trying to do is possible and at the moment I have limited skills but I am learning.

I would like to convert US books to UK versions by searching and replacing US words and phrases to UK. e.g 'color' to 'colour' and 'cell phone' to' mobile phone'.

I would like to do this using two lists so I can add to the lists as I find new words or phrases to convert.

I think the function would be something like:

START
n=0
FETCH LIST#1/STRING#1(color)
FETCH LIST#2/STRING#1(colour)
FIND LIST#1/STRING#1 in the text
IF MATCH FOUND
REPLACE LIST#1/STRING#1 with LIST#2/STRING#1
n+1
GO TO SEARCHn

SEARCHn
FETCH LIST#1/STRING#2
FETCH LIST#2/STRING#2
FIND LIST#1/STRING#2 in the text
IF MATCH FOUND
REPLACE LIST#1/STRING#2 with LIST#2/STRING#2
n+1
GO TO SEARCHn

And so on until end of list.

Is this possible? Has this already been done in some way?
I would be very grateful if someone could point me in the right direction.

Thanks - moldy

[eschwartz]

You'd need to create a single Search that finds ALL words, then the Function-Replace would switch between each word and it's replacement.

You could also do an editor plugin which does each find-replace in turn. (It could take wordlists.)
But no one has written such a calibre editor plugin. In fact, the only publicly posted editor plugin is DiapDealer's plugin for dealing with spans and stuff. Would be nice if people would create more plugins.



Personally, my first step would be to switch the Language metadata to en_GB, activating the en_GB dictionary, and fixing the common misspellings.
That would catch "color" --> "colour" at least, although not "cell phone" --> "mobile phone".

[theducks]

Sigil's Saved Search has a 'chained' (my term) mode where you can make a group of individual saved searches, Then Load the Groupname and execute that

Took me a long time to figure how to use that

Bring up the Saved Search Editor. You will see the controls for this on the Right

[kovidgoyal]

You can create multiple saved search and replaces, select them all and execute them. Give them names that includes something common like [GB], then filter the saved search list by [GB], press Ctrl+A to select all shown searches and execute them.

[theducks]

Originally Posted by kovidgoyal

You can create multiple saved search and replaces, select them all and execute them. Give them names that includes something common like [GB], then filter the saved search list by [GB], press Ctrl+A to select all shown searches and execute them.

Wow
(told you I could be dense)

[moldy]

Hi All and thanks for the helpful replies.

The saved search solution sort of works but is not ideal.

Searching for say 'color' finds color, colorful, coloring etc. which is quite useful. But search and replace 'odor, gives me odour - fine but also deodourant - wrong.
Search and replace 'mom' this gives me mum fine but also mument - wrong.

Trying to use regex to manipulate the search I searched for [Oo]dor[^\w].
This worked but then what about odorless - another search to save and name.

Trying to speed things up I exported the saved searches and opened the file in an old copy of DreamWeaver I have. I could see the search sections so I set up a template to add a search:

{
"case_sensitive": true,
"dot_all": true,
"find": "([Ww])ord([^\w])",
"mode": "regex",
"name": "",
"replace": "\\1ord\\2"
},

I added some words and saved the file but when I tried to import it I got this:

calibre, version 2.41.0
ERROR: Unhandled exception: <b>ValueError</b>:No JSON object could be decoded

calibre 2.41 [64bit] isfrozen: True is64bit: True
Windows-8-6.2.9200 Windows ('64bit', 'WindowsPE')
('Windows', '8', '6.2.9200')
Python 2.7.9
Windows: ('8', '6.2.9200', '', 'Multiprocessor Free')
Successfully initialized third party plugins: Kindle Collections && Diaps Editing Toolbag && Manage Series && Count Pages
Traceback (most recent call last):
File "site-packages\calibre\gui2\tweak_book\search.py", line 1090, in import_searches
File "json\__init__.py", line 338, in loads
File "json\decoder.py", line 366, in decode
File "json\decoder.py", line 384, in raw_decode
ValueError: No JSON object could be decoded

Well thats where I'm up to in my project. As much as I would like to; I don't have the capability to write a plug in.

Any input would be welcomed and much appreciated. Thanks - moldy

[moldy]

Small error in my previous post:

"name": "Word",

I did actually name all my saved searches.

moldy

[HarryT]

Originally Posted by moldy

Small error in my previous post:

"name": "Word",

I did actually name all my saved searches.

moldy

If you make a mistake in a post, you can click the "Edit" button underneath the message to edit it.

[moldy]

Thanks for the tip Harry. I'll remember that for future posts ;-)

[kovidgoyal]

Use the \b operator to match word boundaries, like this

\bcolor\b

And you can create a single saved search to do all your replacing if you are willing to use function mode, something like this

\bcolor|odor|vapor|...\b

and define a function to do the actual replacement,

http://manual.calibre-ebook.com/function_mode.html

def replace(match, number, file_name, metadata, dictionaries, data, functions, *args, **kwargs):
   return {'color':'colour', 'odor':odour', 'vapor':vapour'}.get(match.group(), match.group())

[moldy]

Thanks for your help Kovid. I hadn't heard of the \b boundary before and this is very useful.

I found I can edit the saved searches file in notepad++ and then reload it without problem.

There are a at least a couple of thousand words spelled differently in US v. GB English and of course they have capitalisations and plurals.

I am going to save the words I come across in common usage as individual saved searches and add to the file as I go along.
I will run the searches at the start of an editing session and then use the spell checker to find any words not in my file.
At the moment I am ignoring words that have double letters in GB but not in US - too many of them.

(Incidentally the words check and checker are a problem because in GB we have cheque and chequer as well as check and checker - can't think of a solution!)

My general find-replace is now:

\b([Oo])dor([s]?)\b
\1dour\2

This seems to work OK.

Once I have built up a sizable list of words I will post the file somewhere for anyone to use.

Thanks again - moldy

[Phssthpok]

Originally Posted by moldy

I would like to convert US books to UK versions by searching and replacing US words and phrases to UK. e.g 'color' to 'colour' and 'cell phone' to' mobile phone'.

The problem is fairly intractable -- see for example http://home.comcast.net/~helenajole/Harry.html which gives a textual comparison between "Harry Potter and the Philosopher's Stone" (UK edition) and "Harry Potter and the Sorceror's Stone" (US edition -- they seemed to assume that no-one in the US would be familiar with the concept "Philosopher's Stone" and what the hell is a philosopher anyway? or something.) Changes include "bangs" (US) instead of "fringe" (UK). What are you going to do if someone lets off fireworks and there are "lots of bangs" -- will you have "lots of fringe"?

Similarly, Terry Pratchett's Bromeliad trilogy features a bulldozer called Jukub (JCB) in the UK edition, which is called Cat in the US edition, and all the jokes were changed to suit.

And don't even get me started on common cultural oopsies like "Will you knock me up in the morning?" or "Can I bum a fag?"...

[HarryT]

This whole thing seems like an exercise in futility. As the previous poster says, there are likely to be numerous differences between the UK and US edition of a book besides simple spelling changes. I'd simply accept that American English and British English are different variants of the language.

[Phssthpok]

Originally Posted by Phssthpok

see for example http://home.comcast.net/~helenajole/Harry.html which gives a textual comparison between "Harry Potter and the Philosopher's Stone" (UK edition) and "Harry Potter and the Sorceror's Stone" (US edition)

One other example from this, just for the sake of entertainment:

UK: Dudley had a tantrum because his knickerbocker glory wasn't big enough...

US: Dudley had a tantrum because his knickerbocker glory didn't have enough ice cream on top...

Presumably the concept of anyone serving a not-big-enough knickerbocker glory in the US just did not compute...

[Phssthpok]

Originally Posted by HarryT

I'd simply accept that American English and British English are different variants of the language.

I think it was Oscar Wilde who said that Britain and America are two countries separated by a common language...