Topics

Feature when reading text with dashes


Arno Schuh
 

Hi,

in a lot of HTML newsletters I have an annoying problem: They uses lines with dashes.
NVDA reads dash dash dash dash dash dash dash etc.
-
-
-
-
-
-

Would be nice, if NVDA could recognize a row of lines with the same characters like it do when the same characters are into a single line.
------

Yours sincerely

Arno


Arno Schuh
 

uploaded an example at.

http://www.schmusilin.de/example.htm

Arno Schuh wrote:

Hi,
in a lot of HTML newsletters I have an annoying problem: They uses
lines with dashes.
NVDA reads dash dash dash dash dash dash dash etc.
-
-
-
-
-
-
Would be nice, if NVDA could recognize a row of lines with the same
characters like it do when the same characters are into a single line.
------
Yours sincerely
Arno


Ralf Kefferpuetz
 

Hello Armo,
I can't see any annoying dashes on that page. Where exactly do you see them? Please give me a phrase to search for...

Cheers,
Ralf

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Arno Schuh
Sent: Freitag, 16. August 2019 11:30
To: nvda@groups.io
Subject: [nvda] Fw: Feature when reading text with dashes

uploaded an example at.

http://www.schmusilin.de/example.htm

Arno Schuh wrote:
Hi,

in a lot of HTML newsletters I have an annoying problem: They uses
lines with dashes.
NVDA reads dash dash dash dash dash dash dash etc.
-
-
-
-
-
-

Would be nice, if NVDA could recognize a row of lines with the same
characters like it do when the same characters are into a single line.
------

Yours sincerely

Arno


George Bell <george@...>
 

Goodness knows what the original source came from, but here's line 5.

No wonder it's giving a screen reader the fits!

George

line-height: 0px !important; font-size: 0px !important;">Keine Versandkosten | Kein Mindestbestellwert<div style="display: none; max-height: 0px; overflow: hidden;">&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;&nbsp;&zwnj;</div><br /><img src="http://news.genius.tv/tas9xof7-ak7aug0z-uvyzjypu-11h.gif"

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Ralf Kefferpuetz
Sent: 16 August 2019 12:31
To: nvda@nvda.groups.io
Subject: Re: [nvda] Fw: Feature when reading text with dashes

Hello Armo,
I can't see any annoying dashes on that page. Where exactly do you see them? Please give me a phrase to search for...

Cheers,
Ralf

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Arno Schuh
Sent: Freitag, 16. August 2019 11:30
To: nvda@groups.io
Subject: [nvda] Fw: Feature when reading text with dashes

uploaded an example at.

http://www.schmusilin.de/example.htm

Arno Schuh wrote:
Hi,

in a lot of HTML newsletters I have an annoying problem: They uses
lines with dashes.
NVDA reads dash dash dash dash dash dash dash etc.
-
-
-
-
-
-

Would be nice, if NVDA could recognize a row of lines with the same
characters like it do when the same characters are into a single line.
------

Yours sincerely

Arno


Arno Schuh
 

Hi Ralf,

strange. The newsletter begins with 4 dashes. Do you can't see that?
If you only use speech - probably your speech engine read it aloud.
I use Nuance Vocalizer Steffi as my default German voice.
Probably when useing SAPI voices, Easpeak etc. they didn't do the same?

Yours sincerely

Arno

Ralf Kefferpuetz wrote:

Hello Armo,
I can't see any annoying dashes on that page. Where exactly do you
see them? Please give me a phrase to search for... Cheers,
Ralf
-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Arno
Schuh Sent: Freitag, 16. August 2019 11:30
To: nvda@groups.io
Subject: [nvda] Fw: Feature when reading text with dashes
uploaded an example at.
http://www.schmusilin.de/example.htm
Arno Schuh wrote:
Hi,
in a lot of HTML newsletters I have an annoying problem: They uses
lines with dashes.
NVDA reads dash dash dash dash dash dash dash etc.
-
-
-
-
-
-
Would be nice, if NVDA could recognize a row of lines with the same
characters like it do when the same characters are into a single
line. ------
Yours sincerely
Arno


 

It is a simple matter to use the default dictionary to deal with this if you have material that you read frequently that uses separator lines, regardless of the character used.
 
I long ago added an entry to the speech dictionary to handle lines of repeated asterisks, dashes, or plus signs, when there are more than three in a row.  I've extracted the component parts here:
 
Comment:  Announce any long string of hyphens, plus signs, or asterisks as "separator line"
 
Pattern:  (\+|-|\*){3,}
 
Replacement:  Separator Line
 
Type:   Regular Expression
 
If that is included in the speech dictionary any line consisting of 3 or more of those characters, in any combination, will be announced as "Separator Line."   One could, of course, elect to have no replacement, but I never do that.  I want to know that something has been encountered, and what the nature of it was.
 
I'll include a couple of examples, with a description of what follows before each.  They will all read as "separator line."
 
All Hyphens:
-----------
 
All Asterisks:
**********
 
All Plus Signs:
++++++++++
 
All three characters in repeating combination:
-*+-*+-*+-*+-*+-*+-*+-*+-*+-*+
 
 
 
 


Arno Schuh
 

Seems you haven't read the problem carefully. The dashes are not in a line, but just one dash in a line, but several lines with such a single dash.


Brian Vogel wrote:

It is a simple matter to use the default dictionary to deal with this
if you have material that you read frequently that uses separator
lines, regardless of the character used.

I long ago added an entry to the speech dictionary to handle lines of
repeated asterisks, dashes, or plus signs, when there are more than
three in a row. I've extracted the component parts here:

Comment: Announce any long string of hyphens, plus signs, or
asterisks as "separator line"

Pattern: (\+|-|\*){3,}

Replacement: Separator Line

Type: Regular Expression

If that is included in the speech dictionary any line consisting of 3
or more of those characters, in any combination, will be announced as
"Separator Line." One could, of course, elect to have no replacement,
but I never do that. I want to know that something has been
encountered, and what the nature of it was.

I'll include a couple of examples, with a description of what follows
before each. They will all read as "separator line."

All Hyphens:
-----------

All Asterisks:
**********

All Plus Signs:
++++++++++

All three characters in repeating combination:


 

On Fri, Aug 16, 2019 at 03:57 PM, Arno Schuh wrote:
Seems you haven't read the problem carefully.
And you've not expressed it well.   The webpage provided (which is what I checked) looks like any other webpage, a mix of graphics and text.

You can, easily, eliminate the speaking of lines that are constituted of nothing but a single dash as well.   It's a matter of having a pattern that is not easily confused with anything else.

If what you're concerned about is a line composed of but a single hyphen (or even multiple hyphens, but only hyphens) not being read, use the pattern:

(-)+

as a regular expression match in the default dictionary, then have the replacement be empty.


 
--

Brian - Windows 10 Pro, 64-Bit, Version 1903, Build 18362  

The color of truth is grey.

           ~ André Gide

 

 


Ralf Kefferpuetz
 

Arno,
I don't hear them, with espeak, eloquence and core voicesoeven with symbol level set to all.

Cheers,
Ralf

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Arno Schuh
Sent: Freitag, 16. August 2019 17:48
To: nvda@nvda.groups.io
Subject: Re: [nvda] Fw: Feature when reading text with dashes

Hi Ralf,

strange. The newsletter begins with 4 dashes. Do you can't see that?
If you only use speech - probably your speech engine read it aloud.
I use Nuance Vocalizer Steffi as my default German voice.
Probably when useing SAPI voices, Easpeak etc. they didn't do the same?

Yours sincerely

Arno

Ralf Kefferpuetz wrote:
Hello Armo,
I can't see any annoying dashes on that page. Where exactly do you see
them? Please give me a phrase to search for...

Cheers,
Ralf

-----Original Message-----
From: nvda@nvda.groups.io <nvda@nvda.groups.io> On Behalf Of Arno
Schuh
Sent: Freitag, 16. August 2019 11:30
To: nvda@groups.io
Subject: [nvda] Fw: Feature when reading text with dashes

uploaded an example at.

http://www.schmusilin.de/example.htm

Arno Schuh wrote:
Hi,

in a lot of HTML newsletters I have an annoying problem: They uses
lines with dashes.
NVDA reads dash dash dash dash dash dash dash etc.
-
-
-
-
-
-

Would be nice, if NVDA could recognize a row of lines with the same
characters like it do when the same characters are into a single
line. ------

Yours sincerely

Arno