1. Forum
  2. Usenet
  3. COMP.SYS.ACORN.NETWORKING

  • Browse html special chars

    From Ronald@21:1/5 to All on Sun Dec 25 12:13:41 2022
    I have been setting up html readme's so Browse can be used.
    I have found that while double quote " works, ’ does not.
    A fix is to use $#34; instead.
    Almost sounds like a fault in Browse though.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Harriet Bazley@21:1/5 to Ronald on Mon Dec 26 14:16:37 2022
    On 24 Dec 2022 as I do recall,
    Ronald wrote:

    I have been setting up html readme's so Browse can be used.
    I have found that while double quote " works, ’ does not.
    A fix is to use $#34; instead.
    Almost sounds like a fault in Browse though.


    Does Browse support other HTML 4 entities?
    (e.g. does “ work?)


    --
    Harriet Bazley == Loyaulte me lie ==

    Questions are a burden to others, but answers are a prison for oneself.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ronald@21:1/5 to Harriet Bazley on Tue Dec 27 04:52:12 2022
    In article <14e7e05c5a.harriet@bazleyfamily.co.uk>,
    Harriet Bazley <harriet@bazleyfamily.co.uk> wrote:
    On 24 Dec 2022 as I do recall,
    Ronald wrote:

    I have been setting up html readme's so Browse can be used.
    I have found that while double quote &quot; works, &rsquo; does not.
    A fix is to use $#34; instead.
    Almost sounds like a fault in Browse though.


    Does Browse support other HTML 4 entities?
    (e.g. does &ldquo; work?)

    Hi Harriet,
    no &ldquo; doesnt work.
    A bit lazy on my part, I later found some html xml standards on the
    wiki and it looks like at newer standards, more were introduced.
    The old XML1 only had double quote and &apos; I was supprised that
    Browse didn't support that one.
    I guessing Browse had settled on a standard at the time.
    I don't know the Browse source, I guess the swap table would have
    to be increased or adapted to fix the most common occurences.
    &nbsp; is another one.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Sprow@21:1/5 to Ronald on Thu Dec 29 23:05:50 2022
    On Monday, December 26, 2022 at 3:34:47 PM UTC, Ronald wrote:
    In article <14e7e05c5...@bazleyfamily.co.uk>,
    Harriet Bazley <har...@bazleyfamily.co.uk> wrote:
    On 24 Dec 2022 as I do recall,
    Ronald wrote:

    I have been setting up html readme's so Browse can be used.
    I have found that while double quote &quot; works, &rsquo; does not.
    Does Browse support other HTML 4 entities?
    (e.g. does &ldquo; work?)
    no &ldquo; doesnt work.
    A bit lazy on my part, I later found some html xml standards on the
    wiki and it looks like at newer standards, more were introduced.
    The old XML1 only had double quote and &apos; I was supprised that
    Browse didn't support that one.
    I guessing Browse had settled on a standard at the time.

    ldquo (and rsquo and lsquo) are all in the table of known entities for Browse

    https://gitlab.riscosopen.org/RiscOS/Sources/Lib/HTMLLib/-/blob/master/c/display#L573

    It's worth noting however that these are mapped to UCS4 code points so for them to display properly
    a) you'd need to be running with the unicode Font Manager
    b) iff you're running the Unicode build of Browse you'd also need to select *Alphabet UTF8

    This sloppy test

    <html><body>
    &lsquo;Hello&rsquo;
    </body></html>

    renders fine on Browse 2.20 here,
    Sprow.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ronald@21:1/5 to Sprow on Sat Dec 31 13:53:47 2022
    In article <e583f4a2-cd1b-464b-b80b-8667309fdc92n@googlegroups.com>,
    Sprow <news@sprow.co.uk> wrote:

    ldquo (and rsquo and lsquo) are all in the table of known entities for Browse

    https://gitlab.riscosopen.org/RiscOS/Sources/Lib/HTMLLib/-/blob/master/c/display#L573

    It's worth noting however that these are mapped to UCS4 code points so for them to display properly
    a) you'd need to be running with the unicode Font Manager
    b) iff you're running the Unicode build of Browse you'd also need to select *Alphabet UTF8

    This sloppy test

    <html><body>
    &lsquo;Hello&rsquo;
    </body></html>

    renders fine on Browse 2.20 here,
    Sprow.
    I have been loading !Unicode into ram and running it from there at Boot
    time. I have added running Alphabet UTF8 at that point too so hopefully
    wont be forgotten about. !Unicode is needed by !Netsurf and I have
    subversion requiring it.
    Maybe the effects of Alphabet UTF8 are far reaching, probably needed to
    display UTF8 in Edit, and might show up in subversion and other unixlib programs I dont know yet.
    Thanks, Ronald

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Sprow@21:1/5 to Ronald on Sat Dec 31 00:22:51 2022
    On Saturday, December 31, 2022 at 12:36:11 AM UTC, Ronald wrote:
    In article <e583f4a2-cd1b-464b...@googlegroups.com>,
    Sprow <ne...@sprow.co.uk> wrote:
    ldquo (and rsquo and lsquo) are all in the table of known entities for Browse [and]

    <html><body>
    &lsquo;Hello&rsquo;
    </body></html>

    renders fine on Browse 2.20 here,

    I have been loading !Unicode into ram and running it from there at Boot
    time. I have added running Alphabet UTF8 at that point too so hopefully
    wont be forgotten about. !Unicode is needed by !Netsurf and I have
    subversion requiring it.

    !Unicode is a passive resource folder, you don't need to "load" it as such, it just needs to set up a few system variables which tell stuff where various tables are.

    So what did my 'Hello' example above look like? I can think of 3 outcomes:
    1. Just says Hello with no quotes => does the font you chose have those characters?
    2. Says 'Hello' => great
    3. Says AE_HelloAe_ or similar => those are UTF8 encodings, you need to set the alphabet

    Setting the alphabet is immediate, ie. if you have situation (3) in Latin1 and change to UTF8, when the screen is redrawn Browse will correct the display,
    Sprow.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ronald@21:1/5 to Sprow on Sun Jan 1 22:58:03 2023
    In article <0f9cd186-c334-4d26-a5f1-0e14c8da526bn@googlegroups.com>,
    Sprow <news@sprow.co.uk> wrote:
    On Saturday, December 31, 2022 at 12:36:11 AM UTC, Ronald wrote:
    In article <e583f4a2-cd1b-464b...@googlegroups.com>,
    Sprow <ne...@sprow.co.uk> wrote:
    ldquo (and rsquo and lsquo) are all in the table of known entities for Browse [and]

    <html><body>
    &lsquo;Hello&rsquo;
    </body></html>

    renders fine on Browse 2.20 here,

    I have been loading !Unicode into ram and running it from there at Boot time. I have added running Alphabet UTF8 at that point too so hopefully wont be forgotten about. !Unicode is needed by !Netsurf and I have subversion requiring it.

    !Unicode is a passive resource folder, you don't need to "load" it as such, it
    just needs to set up a few system variables which tell stuff where various tables are.

    Oh, so it is not accessed regularily then, obvious once at start up of a program.


    So what did my 'Hello' example above look like? I can think of 3 outcomes:
    1. Just says Hello with no quotes => does the font you chose have those characters?
    2. Says 'Hello' => great
    3. Says AE_HelloAe_ or similar => those are UTF8 encodings, you need to set the alphabet

    Setting the alphabet is immediate, ie. if you have situation (3) in Latin1 and change to
    UTF8, when the screen is redrawn Browse will correct the display,
    Sprow.

    The difference was noticeable on sites so could see it was working.
    Yes I got a quoted hello from your example.

    At one time, ArcWeb was the fast browser.
    It must have had issues with utf8, there would have been simpler web sites around then.
    I noticed Browse ca handle the redirections involved downloading a Google
    drive link.
    Netsurf wasn't able to lately, I suspect changes at google.
    The other googledrive option was wget.
    Netsurf uses curl which should be superior but somethings up.
    Thanks Ron

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Harriet Bazley@21:1/5 to Ronald on Sun Jan 1 16:38:00 2023
    On 1 Jan 2023 as I do recall,
    Ronald wrote:


    [snip]


    At one time, ArcWeb was the fast browser.
    It must have had issues with utf8, there would have been simpler web sites around then.

    I used to use ArcWeb offline for reading HTML helpfiles etc -- it was
    small and fast, but it didn't support HTML4 entities, which became an increasing problem in terms of readability.

    --
    Harriet Bazley == Loyaulte me lie ==

    The purpose of rules is to make you think before you break them.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Ronald@21:1/5 to Harriet Bazley on Mon Jan 2 21:34:12 2023
    In article <cadc04605a.harriet@bazleyfamily.co.uk>,
    Harriet Bazley <harriet@bazleyfamily.co.uk> wrote:
    On 1 Jan 2023 as I do recall,
    Ronald wrote:


    [snip]


    At one time, ArcWeb was the fast browser.
    It must have had issues with utf8, there would have been simpler web sites around then.

    I used to use ArcWeb offline for reading HTML helpfiles etc -- it was
    small and fast, but it didn't support HTML4 entities, which became an increasing problem in terms of readability.

    It got left behind early on, could be it wasn't OK on StrongArm.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Stewart Brodie@21:1/5 to Ronald on Thu Feb 9 23:06:20 2023
    Ronald <chchron@yahoo.co.nz> wrote:

    In article <cadc04605a.harriet@bazleyfamily.co.uk>,
    Harriet Bazley <harriet@bazleyfamily.co.uk> wrote:
    On 1 Jan 2023 as I do recall,
    Ronald wrote:


    [snip]


    At one time, ArcWeb was the fast browser. It must have had issues with utf8, there would have been simpler web sites around then.

    The finishing touches to the UTF-8 standard were only done in 1994, which post-dates a lot of the development effort.


    I used to use ArcWeb offline for reading HTML helpfiles etc -- it was
    small and fast, but it didn't support HTML4 entities, which became an increasing problem in terms of readability.

    It got left behind early on, could be it wasn't OK on StrongArm.

    It was fine on StrongArm-powered computers - all my computers had SAs.


    --
    Stewart Brodie

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)