Indexing based on a word-list
Moderators: Eyal Redler, redlers, Ori Redler
Indexing based on a word-list
Dear all
Is there any tool in Mellel allowing to create an index based on the word list? It is an extremely convenient feature of Word. It is very rough and there is a lot of work after that of cleansing etc., but my 700 pages book was indexed this way in a few seconds after exporting the text to Word!
But obviously I lost all the formating created in Mellel. So that what can I do to spare me the dozens of hours needed to create one after the other the hundreds of index references in Mellel? Or, alternatively, is there a possibility to export from mellel to another software, create automatically the index references, then reimport to mellel WITH the index references preserved?
Thanks a lot
Etienne
Is there any tool in Mellel allowing to create an index based on the word list? It is an extremely convenient feature of Word. It is very rough and there is a lot of work after that of cleansing etc., but my 700 pages book was indexed this way in a few seconds after exporting the text to Word!
But obviously I lost all the formating created in Mellel. So that what can I do to spare me the dozens of hours needed to create one after the other the hundreds of index references in Mellel? Or, alternatively, is there a possibility to export from mellel to another software, create automatically the index references, then reimport to mellel WITH the index references preserved?
Thanks a lot
Etienne
Re: Indexing based on a word-list
I'm afraid not.Is there any tool in Mellel allowing to create an index based on a word list?
You can also create multiple types of word lists in Nisus, and no cleansing is necessary.It is an extremely convenient feature of Word. It is very rough and there is a lot of work after that of cleansing
You can create a word list in Mellel with a FindSet. You don't need an external application for that.
Create the index in the final word processor.So what can I do to spare me the dozens of hours
However, you can insert an index in Mellel, copy and paste it into another word processor. Thus you get all the indexed items and the corresponding page numbers from the Mellel document (as text), but you won't have any Index Records and Index Marks in the new document.
If you just need the plain index, this may be the way to go. But for this to work, you need to make sure that the page size & font and margin sizes are the same in both documents. Otherwise the page numbers may not match anymore.
Yes. Create the index in the final word processor.Am I right in saying that creating an index in Mellel is a loss of times if we need to export the file to another word-processor?
-
- Knows everything, can prove it
- Posts: 152
- Joined: Fri May 04, 2018 4:00 am
Re: Indexing based on a word-list
Going a little rogue here, and recognizing this isn't a solution for either the faint of heart or the even borderline sane, I see the index terms are specified in tags named index-record in the Mellel xml internals. I don't think it would be hard to extract the list from a Mellel file or import new terms, either.
Yeah, I know, touch internals and enter the test pilot club.
I wrote my own mail merge for Mellel a while back. It's a command line Python script. It takes a CSV file and finds merge fields in the form of {{mergeplaceholder}} in the Mellel document. Then it writes one new Mellel file for each record in the CSV file, merging the text in the expected way.
The merge script hasn't been much use, other than a nice exercise in reading and writing Mellel documents outside of Mellel.
After writing that mail merge, about two hour's work including sniffing out what I needed about Mellel's document format, I feel safe with Mellel's proprietary storage.
Mellel documents are proprietary but might as well be open books. Very easy to interpret.
On a practical note, I usually do mail merges in Affinity Publisher. I use Mellel to prepare text copy for Affinity.
I also have Nisus from before discovering Mellel.
Nisus has a lot going for it. In my view, it's not as stable and consistent as Mellel. It might be true that there's a little bit of a learning curve to parts of Mellel, like how Auto titles or document styles work. On the other hand, once you see how they work they are pretty transparent in operation.
Nisus is the other way around. It looks like a gentle learning curve but it's got quirks. You can say a paragraph style is based on a character style, for example, but changing the underlying character style won't always change the paragraph style.
Of course, if you monkey with the internal format of any product's output it's best to be your own customer support. Bug not the vendor.
Yeah, I know, touch internals and enter the test pilot club.
I wrote my own mail merge for Mellel a while back. It's a command line Python script. It takes a CSV file and finds merge fields in the form of {{mergeplaceholder}} in the Mellel document. Then it writes one new Mellel file for each record in the CSV file, merging the text in the expected way.
The merge script hasn't been much use, other than a nice exercise in reading and writing Mellel documents outside of Mellel.
After writing that mail merge, about two hour's work including sniffing out what I needed about Mellel's document format, I feel safe with Mellel's proprietary storage.
Mellel documents are proprietary but might as well be open books. Very easy to interpret.
On a practical note, I usually do mail merges in Affinity Publisher. I use Mellel to prepare text copy for Affinity.
I also have Nisus from before discovering Mellel.
Nisus has a lot going for it. In my view, it's not as stable and consistent as Mellel. It might be true that there's a little bit of a learning curve to parts of Mellel, like how Auto titles or document styles work. On the other hand, once you see how they work they are pretty transparent in operation.
Nisus is the other way around. It looks like a gentle learning curve but it's got quirks. You can say a paragraph style is based on a character style, for example, but changing the underlying character style won't always change the paragraph style.
Of course, if you monkey with the internal format of any product's output it's best to be your own customer support. Bug not the vendor.
Re: Indexing based on a word-list
How do you get to the Mellel xml internals? Are you using a special xml editor? And if so, which one?
I have not been able to reproduce this behavior. My paragraph style will always change if I change the underlying character style. Tell my how the styles you are using look like, and I will test the configuration.Nisus […] got quirks. You can say a paragraph style is based on a character style, for example, but changing the underlying character style won't always change the paragraph style.
-
- Knows everything, can prove it
- Posts: 152
- Joined: Fri May 04, 2018 4:00 am
Re: Indexing based on a word-list
Hey, Icelander,
I used BBEdit for most of my document explorations. Mellel documents, like Word docx, are zip files.
The simple document construction is one of the things I like about Mellel. Inventing the means to extract ASCII versions of Mellel documents is a short afternoon, even if you start from scratch.
For my purposes, Mellel's format is safe for archival. I don't worry about access to my work.
I used BBEdit for most of my document explorations. Mellel documents, like Word docx, are zip files.
The simple document construction is one of the things I like about Mellel. Inventing the means to extract ASCII versions of Mellel documents is a short afternoon, even if you start from scratch.
For my purposes, Mellel's format is safe for archival. I don't worry about access to my work.
-
- Knows everything, can prove it
- Posts: 152
- Joined: Fri May 04, 2018 4:00 am
Re: Indexing based on a word-list
Almost forgot - the quirks in the brand X word processor.
In that non-Mellel product, I have a document with one character style and two paragraph styes. Paragraph one is based on the character style. Paragraph two is based on paragraph one with font attribute changes.
The mixing of character and paragraph styles, where a paragraph style can be based on a character style, seems to leave loopholes in style definitions. I made three or four adjustments in the character styles. Then, the paragraphs in the document stopped tracking character style changes. I didn't touch the paragraph styles, and I didn't delete, add, or rename styles.
Mellel doesn't get caught in side effects like that, at least not in my experience.
In that non-Mellel product, I have a document with one character style and two paragraph styes. Paragraph one is based on the character style. Paragraph two is based on paragraph one with font attribute changes.
The mixing of character and paragraph styles, where a paragraph style can be based on a character style, seems to leave loopholes in style definitions. I made three or four adjustments in the character styles. Then, the paragraphs in the document stopped tracking character style changes. I didn't touch the paragraph styles, and I didn't delete, add, or rename styles.
Mellel doesn't get caught in side effects like that, at least not in my experience.
Re: Indexing based on a word-list
And exporting the document directly from within Mellel as Plain Text takes even less time. :–)Inventing the means to extract ASCII versions of Mellel documents is a short afternoon, even if you start from scratch.
Why choose a thorny, long-winding serpentine country lane when the broad and comfortable six-lane highway lies open in front of us?
Re: Indexing based on a word-list
I followed your description meticulously, made four adjustments in the character style, and everything in the paragraph styles followed suit as expected.Amontillado wrote: ↑Thu Sep 01, 2022 12:08 am In that non-Mellel product, I have a document with one character style and two paragraph styes. Paragraph one is based on the character style. Paragraph two is based on paragraph one with font attribute changes. […] I made three or four adjustments in the character styles. Then, the paragraphs in the document stopped tracking character style changes.
Try this:
1. Save the document and restart Nisus. Does the problem persist?
2. If so, turn off you machine and make a restart after several minutes.
3. If the problem persists, try to "Update All Stale Content" in Tools > Automatic Content > Update All Stale Content
4. This can be a bug related to your macOS. I'm running High Sierra, and everything is OK here.
If you are interested in further troubleshooting, you must tell me what adjustments you did exactly so that I can mirror them on my machine.
-
- Knows everything, can prove it
- Posts: 152
- Joined: Fri May 04, 2018 4:00 am
Re: Indexing based on a word-list
Right, but if your descendants find your Mellel files on an old backup 40 years from now, even if Mellel is no longer available the text is accessible with very little investigation.Icelander wrote: ↑Thu Sep 01, 2022 10:35 amAnd exporting the document directly from within Mellel as Plain Text takes even less time. :–)Inventing the means to extract ASCII versions of Mellel documents is a short afternoon, even if you start from scratch.
Why choose a thorny, long-winding serpentine country lane when the broad and comfortable six-lane highway lies open in front of us?
-
- Knows everything, can prove it
- Posts: 152
- Joined: Fri May 04, 2018 4:00 am
Re: Indexing based on a word-list
I appreciate your help, but out of deference to our host it's probably out of scope for this venue. I really appreciate your offer. In general, I have had the expected results. Twice I've had unexpected results. Nothing's perfect, I'm not saying anything bad about the other product.Icelander wrote: ↑Thu Sep 01, 2022 10:39 amI followed your description meticulously, made four adjustments in the character style, and everything in the paragraph styles followed suit as expected.Amontillado wrote: ↑Thu Sep 01, 2022 12:08 am In that non-Mellel product, I have a document with one character style and two paragraph styes. Paragraph one is based on the character style. Paragraph two is based on paragraph one with font attribute changes. […] I made three or four adjustments in the character styles. Then, the paragraphs in the document stopped tracking character style changes.
Try this:
1. Save the document and restart Nisus. Does the problem persist?
2. If so, turn off you machine and make a restart after several minutes.
3. If the problem persists, try to "Update All Stale Content" in Tools > Automatic Content > Update All Stale Content
4. This can be a bug related to your macOS. I'm running High Sierra, and everything is OK here.
If you are interested in further troubleshooting, you must tell me what adjustments you did exactly so that I can mirror them on my machine.
-
- Co-founder
- Posts: 695
- Joined: Thu Oct 27, 2005 9:15 am
Re: Indexing based on a word-list
There is no way to create an index based on a word list. We plan to add that feature soon.
While on the subject, what kind of export options regarding index would you like to see? (everybody is invited to comment)
While on the subject, what kind of export options regarding index would you like to see? (everybody is invited to comment)
Eyal Redler
----------------------
Co-Founder and Owner at Mellel
Facebook: http://www.facebook.com/mellelwordprocessor
YouTube: http://www.youtube.com/user/MellelRedlex
Donate: https://www.paypal.com/donate/?hosted_b ... 2LWB33YBZW
----------------------
Co-Founder and Owner at Mellel
Facebook: http://www.facebook.com/mellelwordprocessor
YouTube: http://www.youtube.com/user/MellelRedlex
Donate: https://www.paypal.com/donate/?hosted_b ... 2LWB33YBZW
-
- Knows everything, can prove it
- Posts: 152
- Joined: Fri May 04, 2018 4:00 am
Re: Indexing based on a word-list
It looks like there aren't many products that support index term import or export. Some solutions are based on scripts in macro languages.
Exporting and importing indexes is already possible, at least in limited ways.
Use File->Duplicate, Edit->Select all in the new copy, and delete all the content. Now you have your index terms from the first document "exported" to a new file.
If you want to "import" index terms, that's possible too. Create a new file with the index terms from your first document. Globalize the style set in the second document you want to import the index terms to. Select the globalized style set in the empty document. Select all, copy, and paste to get the content from the second file into a document with the index terms from the first file. De-globalize the style set if desired (delete it, let it live on in the individual documents).
That could be more convenient, of course.
A menu command to export index terms to an empty Mellel document would cover the export in one step. Another menu command to merge the indexes from another document would cover the import step, and would allow adding to an index already started. Duplicate terms should be ignored if they are within indexes of the same name.
For more universal interface, a unicode file with one term per line would be a nice import/export format. Indentation could assign grouping. Special leading characters, like hash or bang, on indented lines could signify direct or generic references. Remarks, tags, and marker color could be similarly signified.
Or XML import and export. If Mellel could both import and export XML index term files it would also serve to edit them. Import, modify the terms in the side pane, and export.
Fun features to think about. I've included an index with maybe three documents in my life, so count all this as no more than frivolous speculation.
It would be a cool feature, though.
Exporting and importing indexes is already possible, at least in limited ways.
Use File->Duplicate, Edit->Select all in the new copy, and delete all the content. Now you have your index terms from the first document "exported" to a new file.
If you want to "import" index terms, that's possible too. Create a new file with the index terms from your first document. Globalize the style set in the second document you want to import the index terms to. Select the globalized style set in the empty document. Select all, copy, and paste to get the content from the second file into a document with the index terms from the first file. De-globalize the style set if desired (delete it, let it live on in the individual documents).
That could be more convenient, of course.
A menu command to export index terms to an empty Mellel document would cover the export in one step. Another menu command to merge the indexes from another document would cover the import step, and would allow adding to an index already started. Duplicate terms should be ignored if they are within indexes of the same name.
For more universal interface, a unicode file with one term per line would be a nice import/export format. Indentation could assign grouping. Special leading characters, like hash or bang, on indented lines could signify direct or generic references. Remarks, tags, and marker color could be similarly signified.
Or XML import and export. If Mellel could both import and export XML index term files it would also serve to edit them. Import, modify the terms in the side pane, and export.
Fun features to think about. I've included an index with maybe three documents in my life, so count all this as no more than frivolous speculation.
It would be a cool feature, though.
Re: Indexing based on a word-list
Dear Eyal
thanks for your answer. It might be useful to be able to create the index records in Mellel and to be able to export them. That is, if in my text I have marked "Samarcande" as an indexed word in the whole text in Mellel, after the file is exported it should be seen as { XE "Samarcande" } in Word so that the index itself can be created in Word if needed (which is quite often the case with the editors)
yours
Etienne
PS a very small side problem: in the French localization in
Insert-> Index -> Quick Index popover-> Add record : you cannot deferentiate between the 2 choices on the right (the buttons are too small so that you can only read in both of them "ajouter enreg" = add record, but with no way to know the difference between one another)
thanks for your answer. It might be useful to be able to create the index records in Mellel and to be able to export them. That is, if in my text I have marked "Samarcande" as an indexed word in the whole text in Mellel, after the file is exported it should be seen as { XE "Samarcande" } in Word so that the index itself can be created in Word if needed (which is quite often the case with the editors)
yours
Etienne
PS a very small side problem: in the French localization in
Insert-> Index -> Quick Index popover-> Add record : you cannot deferentiate between the 2 choices on the right (the buttons are too small so that you can only read in both of them "ajouter enreg" = add record, but with no way to know the difference between one another)
-
- Co-founder
- Posts: 695
- Joined: Thu Oct 27, 2005 9:15 am
Re: Indexing based on a word-list
Thanks for elaborating on this.
Regarding export, note that if you export to docx, index marks are exported into msword index fields ({ XE "Samarcande" })
Regarding export, note that if you export to docx, index marks are exported into msword index fields ({ XE "Samarcande" })
Eyal Redler
----------------------
Co-Founder and Owner at Mellel
Facebook: http://www.facebook.com/mellelwordprocessor
YouTube: http://www.youtube.com/user/MellelRedlex
Donate: https://www.paypal.com/donate/?hosted_b ... 2LWB33YBZW
----------------------
Co-Founder and Owner at Mellel
Facebook: http://www.facebook.com/mellelwordprocessor
YouTube: http://www.youtube.com/user/MellelRedlex
Donate: https://www.paypal.com/donate/?hosted_b ... 2LWB33YBZW