On Wed, 2006-11-08 at 15:19 +0100, [EMAIL PROTECTED] wrote: > On Wed, 8 Nov 2006, Lars Clausen wrote: > > > > It certainly is way more than I'd like... it ends up being 36 pages. > > > Anyway, I thought I send it to you somehow, any preferences? > > > > Mail it to me at [EMAIL PROTECTED], not to the mailing list. > > I used my head and simply put it here: > > http://www.md.kth.se/~chr/dia > > so you or anyone else on this list can simply get it. Oh, the drawing now > uses A4-pages instead of Letter. > > > Did you end up doing it manually? > > Yes :-( > > I had to tweak quite a bit to get the shapes to fit within a small set of > pages, so not all shapes are in their "original" size. Btw, I wouldn't > recommend doing this again... it's quite a bit of work. > > If you were to do it manually again, I strongly recommend doing one cheat > sheet for each category of shapes. If Dia can included other .dia-files, > it'll be quite easy to then create a composite document with all the > pages. > > It's probably much more practical to automatically generate a document > like this. Things might be simplified if we create one document for each > category of shapes though. > > But above all, what I miss are explanations about the shapes. It's quite > difficult to look at a cheat sheet and see that this block actually > accepts several arguments that let you attach text to it.
Yes, not all shape creates are equally good about describing the shapes. In fact, very few are good about it at all. > As an aside, I was pleasantly surprised to find zig-zag lines that you can > attach text to. That's something I've been looking for:-) > > A few comments: A few? :) > > * Being able to automatically generate these documents, and also place the > name of the shape below would be really useful. Maybe even genereate > is a web page? Maybe a python plugin? Hans? > * Something that was quite annoying is that some shapes take up a much > larger area than they appear to do on screen. Something that happened > frequently was that I got one or two extra blank pages, simply because a > shape extended invisibly to the next page. The connection point is > a good example of this. A very small thing on screen compared to the > extent it has. Which shapes do this? > ** Maybe it'd be nice if there was an option to display the "extent" of > all the shapes? (Or maybe the shapes should be changed to not have > such large extents). There's a secret setting render_bounding_boxes (changeable in ~/.dia/persistence that turns such rendering on. > * In one case Dia reported a certain number of pages, but when printing > to Adobe PDF I actually got more pages. There can be some issues when an object is on the edge, and the width of a line may or may not get counted. Did the PS file have more pages, too? > * Why are the shapes so big by default? Take the 'circuit shapes' for > instance, I can't imagine that anyone actually use them in that size? Ancient default. It's a size that at 100% makes it reasonable to handle. > * I really missed a method for aligning objects in an array format. > Something where you select a bunch of objects, click 'aling in array' > and then enter 3x5 (row x column) and the objects are aligned > accordingly. Yeah, using align vertically/horizontally in subgroups gets tiring after a while. There's nothing basically wrong with such an idea, it just takes a bit of implementation to do the dialog -- which I have no time for. > * I got fooled by the 'align spread out horizontaly', as it takes all the > selected objects and aligns them. Since I'd selected objects on > different "rows", they were all spread out instead of this being done > "row-wise". (I realize that some kind of threshold value is needed in > order to determine which objects belong to a row). I'd rather keep those align functions simple and add an array align than try to guess at what the user wants and ending up with the user having to guess at what happens. > * Is there some way to figure out what the keyboard shortcuts are for > the different commands? (Assuming that commands without a shown > keyboard shortcut actually has one). If they're not shown, they don't have one. > Alternatively, is there a way to assign your own keyboard shortcuts? Yes, but it's turned off in GTK by default. Apparently too many people set up shortcuts by accident and got confused. I don't remember the way to set it offhand, but it's probably in gconf. > * I'd prefer being able to move objects around on screen using the > keyboard in certain cases. (Other drawing software lets you do this > using the arrow keys for instance. Is this possible in Dia?). Not yet. It's been mentioned many times, and I did take a look at it once but didn't quite far and removed the initial crud from the code. It's a little tricky as long as text input is "always on" and also uses the arrow keys. Once more, proper text input is a requirement for other features. > > > One thing that really bugged me was that sometimes things didn't quite > > > fit on a page, so a whole row or column of additional pages were added > > > in the printout. It'd be really good if Dia could somehow show which > > > pages actually have some content on them. > > > > That's an interesting idea. Have the content-less 'pages' greyed out or > > something. Only problem is, the pages are fully virtual, just lines on > > the canvas. Figuring out which pages are empty would require running a > > non-trivial amount of code -- though it could be made to work. > > Maybe there could be a "preview"-command or "preview-mode" where pages > that will be printed are shown differently? (Thicker line?) > > Another option would be to have the code checking what pages that will be > printed to be exectued when using the command 'Redraw'. (There should be a > keyboard shortcut for that command btw). The redraw command should not be necessary, as in bug-free code it should redraw appropriately automatically. Greying out outside a rectangular area would be doable when doing the page break lines anyway, but greying out "corner" empty pages would be harder. > > > In addition, I frequently used the File->Page setup menu to find out how > > > many pages Dia thought was used. That information might be nice to have > > > in > > > the status bar. > > > > > > In addition, it seems that the page count isn't always updated when > > > turning the display of a layer on/off. > > > > That appears to be a bug. Seems like changing the visibility of layers > > should force a recompute but doesn't. If you try to make a layer > > invisible, then change something in the diagram, you can see the update. > > Yes, I did notice that if I moved something then the page count was > updated. I think 'Redraw' should force a recount, but it doesn't do that > today. That bug is now fixed in CVS, and visibility is now even undoable. -Lars _______________________________________________ Dia-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/dia-list FAQ at http://www.gnome.org/projects/dia/faq.html Main page at http://www.gnome.org/projects/dia
