Text flow around image is incorrect (Scribus 1.5.2)

Previous topic - Next topic

AdeSmith

Hey all. I've just started a new project and both times so far where I've added art and set it to have the text flow around it the first 3 to 4 letters of the word below sits 1 line above the others. No matter what I do to the bounding box I cannot seem to get the text to line up. Any ideas why this is happening?

The solution I've come up with so far is to create a new text frame below and flow the text into that rather than have the image determine where the text should be.


GarryP

Can you supply some reference materials please? (Best to use "Collect for Output" to generate a file with everything in it.) It's very difficult to know what's happening otherwise.

AdeSmith

#2
I'm not quite sure how to give you the info as I'm kinda new to Scribus. If I can get a heads up on how to do so I will get you the info. I have a screen capture of what is happening added below.The 1 in the date under the picture is always on the line above the rest of the text. I've managed to add a space in it's place and this has allowed it to drop in line with the rest but it's still not working correctly.

[attachment deleted by admin]

Laser

Hi AdeSmith!


I'm still not sure of your problem.
I see an imageframe - not fitted to the image. Is that right?
And I see one text frame - are you working with rows?

Best would be to collect for output and upload the zip.
And could you show how it should look like - perhaps with Gimp. No need to be perfect, just to understand.


Thanks and regards
Laser

Andreas S

I noticed this effect many times - it's a bit difficult to overcome and looks like a bug to me. If I remember right this behaviour is in Scribus at least since 1.5.0.

When the text flows around an image (even with a horizontal baseline), then the first letters of text appear directly below the image border, and with the first letter with a cap height Scribus starts a new line as the cap height doesn't fit in the available space.

GarryP

1. My first recommendation, for people learning Scribus, would be to use the current stable version - currently 1.4.6 - rather than the latest development version. Development versions - while being mostly usable - can contain "hidden" bugs from changes or new features that haven't yet been properly integrated into the code. This can lead to the development software doing strange things that the stable version doesn't.

Also, development versions may contain features that don't make it into later stable versions of the software. In these cases it might mean that your documents can't be loaded into future versions of the software that don't contain those features. You would have to maintain multiple versions of Scribus while remembering which version to use on different documents.

Bottom line: Unless you're specifically doing development testing, and/or really know what you're doing, use the current stable version instead of a development version.

2. My second recommendation, for people wanting answers to problems, would be to give the forum members as much information about the problem as possible. I realise that it's difficult to know what information to give as you might not know what's relevant but, at the very least, let people see everything that you can see.

If you're providing a screen shot then it's best to give the whole screen - or failing that, the whole Scribus window. Just a portion of the screen may not give enough information. For instance, in the provided screen shot it looks like there's a two-column text frame being used and there looks to be something odd at the start of the second column but we can't see what's at the end of the first column. This means that we don't know what the full context of the problem is. In this instance, what's at the end of the first column may be important to figure out what's going on in the second column. It might not but we can't tell. Taking a screen shot of everything is as easy as it is for just a part of the screen so why restrict the information given?

Bottom line: If you want help then help the people you're asking help from by giving them as much help as you can.

Note: If your screen shot contains sensitive information that you don't want other people to see then you can use something like GIMP to quickly blur sections of the image.

3. The best way to give the forum members as much information as possible is to use the "File -> Collect For Output" function. By using this function you can create a new folder which will contain a copy of the document, copies of all of the images used, and - if you use the right check boxes - copies of all of the fonts and colour management information used in the document.

Once this new folder has been created you can ZIP it and post the ZIP on the forum (as long as the size of the ZIP doesn't break the forum limits, in which case you could add a link to some cloud storage instead). This provides people with a neat package of everything that you're using and makes it much easier to see what's going on. Essentially they're working with the same document that you are.

There is, however, one caveat that brings us back to my first point, and that is that if you collect your output from a development version of Scribus then you are restricting the number of people who can help you to only those that have the same development version installed on their machine that you're using.

For instance, if you collect the output from version 1.5.2 then the only people who can properly check your work are people who have 1.5.2 installed. Version 1.5.1 might not work the same way as 1.5.2, or 1.5.0, or any other development versions so whoever tries to find out what the problem is using a different version might not get the same results as you. To add to the problem, there will be many different versions of, for instance, 1.5.2 depending on when different individual's software was compiled and each version will work differently in different ways. This makes it more important that you stick with the stable version unless you really need something from a development version.

Bottom line: If you want someone to be able to really look at a problem properly then collect your output and post a ZIP but be aware that collecting the output from development versions can make things more awkward for people wanting to help.

Note: Be aware of the copyright situation of the images and fonts you collect for output. By collecting for output and posting a ZIP with them in you are essentially distributing them and could be doing so illegally if they have copyright restrictions. Just make sure that everything is okay before posting.

Epilogue. I hope I haven't scared anyone away from using the forum by saying all of this. I've tried to keep the tone light but they are important issues. No-one on the forum gets paid to help. We give our free time to help others so it's nice if people help us by not giving us too much extra work to do.

(AdeSmith, what I've said is relevant to your original post but it's not specifically targeted at yourself. It's just general information for anyone who reads the forum, I'm not "having a dig".)

a.l.e

hi andreas,

can you please share a .sla (collected for output) that shows the issue?


hi garry,

i tend to 100% agree with you about the use of development versions of scribus.

but i'm tired of fighting against the overly enthusiastic release notes issued by the scribus team (that can too easily be misinterpreted by most users):

https://wiki.scribus.net/canvas/1.5.2_Release

all you're saying in your post is very sane, but by keeping on writing such walls of text and explain what should have been obvious from the beginning, i fear (and you fear too) that we are scaring away people...

amen.

AdmFubar

have you tried right clicking on the image and in the pop up menu select "Adjust Frame to Image" ?

If that isnt the cure. I'd take a closer look at the image. It may have section of tranparent pixels along the bottom. A little editing to the image would help then.
Using Scribus 1.6.1, openSUSE 15.5
Advanced hobbyist

solaero

I have this problem occasionally. I usually solve it by adjusting the contour line of the text box until I get the correct flow. Sometimes moving the image helps, too. Mostly it's just trial and error until I get something that looks palatable.

GarryP

AdmFubar, I don't think the problem is with the gap inside the image frame (the gap looks like it has been created to try and fix the problem). It seems to be an issue with how text flows around the frame.

solaero, using the contour line of the image frame - as you suggested - is a much better way of controlling text flow than trying to change the bounding box. In general, the bounding box should be used to say what part of the image is shown and the contour line should be used to control text flow around the image. However, without access to the original SLA it's impossible to know if changing to the contour line would fix the problem as it's been stated.

a.l.e

hey guys,

how can it happen? several people saying "me too" and nobody can share a file that shows the problem?

i've just tried to replicate with a rectangular frame and a matching contour line and i cannot get that effect...

well, if i have to be honest, i can get something similar. and that similarity seems to suggest that the contour line in the screenshot is not correctly defined... but those are just suppositions!

please, share a .sla (if possible collected for output; if possible one single page) that shows the problem!

if you want other people to understand what is happening and eventually get the team to fix the bug, we need something that shows the wrong behavior! (and let replicate it)

GarryP

a.l.e, (sorry, I forgot that you posted earlier) regarding the "wall of text" issue you raised: Yes, sometimes it can be a bit tiresome having to explain things multiple times but sometimes a few words that don't say much more then RTFM isn't enough for beginners. I prefer to be verbose and give a full explanation than not give enough information which is why I sometimes go on a bit too much. I'd hope that was more appreciated by the reader than just giving them a few quick bullet points and expecting them to know what I was talking about.

I've had a - sort of - related idea which I'll put in another post. It might be something interesting to think about.

AdeSmith

Thanks everyone. I've replicated the effect on a new document (the other was too big). The photo is set up with a frame sized bounding box. If you put a space before the word Britannia below it then the text moves down to it's correct position with the rest of the paragraph, otherwise it seems impossible to get the text to flow correctly. Placing the cursor at the start of the word Britannia seems to make it huge, reaching to the top of the page, could this be part of the reason?

I also agree that using a stable version makes more sense, in fact  I've been using 1.4.6 up until this project but the people that supplied me with the template created it in 1.5.2 and obviously that won't open in the one I use regularly. I'll pass on the info to them so that others may not get caught out by bugs in the future.

If i can do anything else to help please let me know.





[attachment deleted by admin]

a.l.e

hey,

this is what i get on the page that is not empty.

it does not match your screenshot but the layout seems to be correct...

[attachment deleted by admin]

Laser

#14
a-l-e: Depends on the font ;-)
I had to replace and took Bitstream Versa Sans Romanand got the effect.

But as I suggested - it's First Line Offset. We just had a bug similiar to this and it has been fixed afair.
(Here it is.)

@AdeSmith: If you change from Maximum Ascent to Font Ascent the problem is solved, right?
Would this be a solution for you?


Rgds
Laser