Diagnosing a Sigil ID nightmare

[LukeA]

Originally Posted by Hitch

Sigil's not a word-processor; it assumes that its users are smart enough to know HTML, XHTML and CSS. {smile}.

Then it should have giant red banner when in book mode stating "USE AT YOUR OWN RISK"

If it looks like a word processor, it should act like a word processor. Why should ePub creation be limited to only those nerds who have wizard-like techy skills? Most nerds can't write and most writers aren't nerds. If what you propose as a purpose for Sigil is correct, it is made for a very small market. We need to make these tools to simplify life for real people, not so that they are used only by people who don't need them and can work with code directly.

[LukeA]

Originally Posted by Hitch

Do you seriously think that Sigil should have generated thousands of new id's for the thousands of paragraphs in that ePUB of 600 printed pages?

The document I'm working with already had an id on (almost) every element when I got it. I don't know what tool generated the document, but there are probably thousands of unique ids. The dupes are from my edits.

If generating an unused id is really problematic then copy while stripping the id would be perfectly fine. If you copy, say, a heading and strip the id, Sigil will generate an id if the heading is included in the ToC. If not and you need an id for a link, Sigil has the dialog to let you do it (though it is then up to you to ensure no duplicates - perhaps prompting with a proposed unique id would be useful). I used a tool that had such an id prompting feature in it once - I think it was a software documentation product a couple of decades ago.

Actually, copy and paste should zap or generate an id; cut and paste should preserve the id on the first paste and either zap or gen on the second and subsequent. Cut and paste is simply a move and a link should not be broken as a result. Subsequent pastes after a move are the same as a copy and the id problem should be addressed for those.

[Turtle91]

Originally Posted by LukeA

Why should ePub creation be limited to only those nerds who have wizard-like techy skills? Most nerds can't write and most writers aren't nerds.

You did NOT just call us NERDS did you!!??!! Dude...no wonder you are having problems with Sigil...if you can't even tell the difference between NERDS and GEEKS!

[Hitch]

Originally Posted by Doitsu

Actually, generating new unique ids for paragraphs isn't that complicated in Sigil. Thanks to the comprehensive Index code implemented by Meme, all you have to do is:

1. Delete all paragraph ids and select Tools > Index > Index Editor.
2. Right-click the Index Editor window and select Autofill.
3. Select Tools > Index > Create Index.

This will cause Sigil to automatically add consecutive ids to each paragraph in the epub. (Not that they'll be particularly useful.)

Doits:

You know I love ya, but, hell...why on earth would someone do that? (I could actually see that being useful if you were doing an outline, all done only with headers, I suppose...)

Hitch

[Hitch]

Originally Posted by LukeA

Then it should have giant red banner when in book mode stating "USE AT YOUR OWN RISK"

If it looks like a word processor, it should act like a word processor. Why should ePub creation be limited to only those nerds who have wizard-like techy skills? Most nerds can't write and most writers aren't nerds. If what you propose as a purpose for Sigil is correct, it is made for a very small market. We need to make these tools to simplify life for real people, not so that they are used only by people who don't need them and can work with code directly.

It's an Open Source program which is, indeed, pretty much clearly marked as "USE AT YOUR OWN RISK" by being labeled as a beta, and occasionally an alpha. If you want an end-user program, buy Scrivener or Jutoh. Each is about $40.00. Sigil is not an authorial tool, as has been discussed in this forum at some length. It's a tool expressly made for those who already know html, xhtml and CSS. Feel free to search on this topic. I'm sorry, but just because to your eyes it looks like a Word-processor doesn't make it a word-processor. You indicated that someone else came to you for help, which says to me that you've professed some wizardy skills yourself, does it not?

Hitch

[LukeA]

Mitch,

I guess I just have a hard time accepting the default notion that open source = toy. I see them as an alternative to commercial, not-necessarily-better software. Apache Web Server is an example - in spite of its headaches (I managed a server farm for several years so I know about the headaches), it is definitely a professional grade product.

My default position is that open source should endeavour to be the best it can be.

BTW - it wasn't that someone asked me to help them - I just picked up a previously worked on project. No skills were asked for or offered. And since I've been retired for a few years, picking up HTML/CSS/regex etc again has been a memory refresher; I didn't realize I'd forgotten so much!

[Hitch]

Originally Posted by LukeA

Mitch,

I guess I just have a hard time accepting the default notion that open source = toy. I see them as an alternative to commercial, not-necessarily-better software. Apache Web Server is an example - in spite of its headaches (I managed a server farm for several years so I know about the headaches), it is definitely a professional grade product.

My default position is that open source should endeavour to be the best it can be.

I never said that Sigil was a toy; please don't miscast what I've said. I said, it was labled as a BETA and occasionally, an ALPHA, which is where the nature of it should have been more clear. Nor did I say it was in any way a "toy;" I said it was a tool for people who already know html, XHTML, and CSS. It's a tool for commercial bookmakers and serious amateur bookmakers. It requires more knowledge to use it, not less. It's neither a word processor nor a toy.

Your expectations of it were simply incorrect--that's all. If you want a product that requires less knowledge of the user, as I stated previously, you can use Jutoh or Scrivener, which are paid products that provide service and customer service for less-knowledgeable users. Sigil is supported by OS donations (some folks here donate quite a bit more for Sigil on an ongoing basis than they would ever pay for a paid product) and does not provide customer service to those folks who don't know what CSS is, or HTML, etc. That's the distinction I was endeavoring to make; not that it was a "toy" in any fashion. Paid products that specialize in allowing non-technological-savvy people to make technological products have the luxury of providing support to those that don't understand the "tech" side. Products like Sigil, or most any other OS-supported endeavor, don't. The crowds to which Sigil versus Scrivener and Jutoh appeal are completely different, and the products serve utterly different purposes. Sigil is not Scrivener, nor does it attempt to be. Sigil is, as I stated, a tool for crafting books using XHTML, HTML and CSS; not a word-processing tool that happens to make eBooks.

Sigil, in the simplest terms I can state it, is not trying to be Scrivener or Jutoh. That's the point. It's not some "lesser cousin," it's a completely different tool. Just as an "FYI," there is and has been considerable discussion given to eliminating BookView entirely; perhaps that will crystallize the difference for you. A large number of us don't use it at all (which is why we now have Preview).

Hopefully, that clears up any confusion.

Hitch (not Mitch).

[mrmikel]

Several of us have been experimenting with Altantis Word Processor which outputs epubs. That may the appropriate tool for someone who doesn't understand HTML and CSS and doesn't want to.

The complaint is that Sigil is not a Swiss Army Knife. Those of us who use it recognize a scalpel when we see it. And like a scalpel, it takes knowledge, because with the Regex Search and Replace you can cause the ebook patient to lose both its heart and its mind, along with your own in the bargain.

[Doitsu]

Originally Posted by Hitch

You know I love ya, but, hell...why on earth would someone do that? (I could actually see that being useful if you were doing an outline, all done only with headers, I suppose...) Hitch

To quote myself:

Originally Posted by Doitsu

...
This will cause Sigil to automatically add consecutive ids to each paragraph in the epub.
(Not that they'll be particularly useful.)

I cannot think of a scenario that'd require adding unique ids to each paragraph, however, since it is theoretically possible that someone might require this feature, I merely wanted to point out that it can be done automatically.

[LukeA]

Originally Posted by Hitch

Your expectations of it were simply incorrect--that's all. If you want a product that requires less knowledge of the user,

No, I want a product that works correctly when used as provided.

I don't have a problem with it providing the ability to go under the covers and fiddle with the code. I do have a problem with the thing producing an incorrect result when used in a manner that it appears to support.

The fix is simple - don't duplicate ids. That shouldn't be a major re-design or massive code change.

The idea that the user is at fault for a simple oversight on the part of the programmers is silly. It's beta - so fine, the problem is identified and should be fixed. The solution is not for all users to be Hairy Chested Real Experts and toil at the lowest level of code. That's just not productive.

Hitch (not Mitch).

Sorry!

[mrmikel]

It may not be as simple as suggested. You are connecting one part, search and replace with establishing and checking links. If this step occurs every time, it will slow the program down. The slowdown will be in direct proportion to the number of links existing.

I believe the main body of users would prefer faster to correct at every stage. In fact there was a modification made to allow it to be not correct, so a work at a certain stage but still not valid could be saved.

[Toxaris]

Originally Posted by LukeA

No, I want a product that works correctly when used as provided.

I don't have a problem with it providing the ability to go under the covers and fiddle with the code. I do have a problem with the thing producing an incorrect result when used in a manner that it appears to support.

The fix is simple - don't duplicate ids. That shouldn't be a major re-design or massive code change.

Use as provided is dangerous statement. Every product can be used differently as designed or though up, even 'professional' products. Sometimes it works, sometimes it doesn't. I can't count the times I have had a developer of a product on the line who said: 'I never imagined it to be used for that' for specific functions or calls.

Quite frankly I would be horrified if I had so many id's, but that is me. I wouldn't call the fix simple. I do not know the code, so I cannot make that assesment. I can image that it can be quite difficult to code correctly, since id's may be called upon. How to decide which links to maintain and which links to delete/change? Especially when things are copied/duplicated. The cure might be worse than the problem.