- entry 2025/10/15

An apology is due

I got a 502 Bad Gateway error (which is why I suspected SNI).
Sprow tried it this morning and got a 421 misdirected request.
I tried it when I came home to capture what the error message for me actually was and...
...it works!?

I should note that I told wget to ignore the certificates as it was complaining and I didn't know if maybe my CAbundle was out of date or something; NetSurf didn't say anything (neither, for that matter, did Firefox on Linux), and according to Sprow at the time he tried the server has the wrong certificate (does not match provided host name); so maybe what was going on was AcornSSL was refusing to make a connection, failing to do anything useful about it (isn't it supposed to open a "what do you want to do now?" dialogue?) and ending up in a muddle?

Whatever, it's now working and nothing has changed here so it's clearly the server that was playing silly buggers in some weird way that caused Acorn's fetcher to malfunction.

tl;dr: Rick shot the messenger (but he didn't shoot the deputy).

Tea status

At the moment Tea is hardwired to the London region. Well, that's the default and there's currently no mechanism to change it (I'll need to add a site fetch when you enter the postcode to translate that into a network ID).

It appears as if the channels no longer have any descriptions, so I'll disable the information that would pop up if you right-click on a channel; there's no point now as there's nothing to show. I won't remove the code just yet, just in case descriptions are added; but it'll probably be a build conditional.

It also looks like the new EPG no longer has any search facility. This... probably won't bother most of you as the search was utterly crap - enter a full name like "Malory Towers" or "The Good Life" and it would report nothing at all; it only supported single words, and even then returned stuff in a random order including in the past.

What will be a little more challenging to deal with is the programme data is not only supplied as a massive dump of everything on a given day (hence why the fetch buffer is huge now), but it also runs very literally that day, as in midnight to midnight. I think I'll keep the current behaviour of splitting at 5am like it used to (as plenty of films and such span midnight); but I'm not sure how to handle this just yet. Probably work out what the logical day is (which may be "yesterday" if you're looking at it at 1am) and fetch what's necessary (day and day+1) to build up the array.
All while ignoring those channels we're not interested in.

This is, of course, why I'm writing a blog post rather than coding. It's going to require massive changes and, well, I don't feel up to it right now. But Tea is reading and parsing the channel list; so we're getting somewhere, slowly.

Oh, and I have identified that the day offsets are C's time_t (seconds since 1970 or somesuch) with the time set to midnight GMT. This isn't a big surprise, as time_t is how the previous code did it, as JavaScript uses the same time system.
And, of course, add/subtract 86400 to change day (which is 24 hours × 60 minutes × 60 seconds).

Now, on the plus side, it looks as if programmes now have genre metadata that appears to broadly match this list (dvb.org), though the EPG references 2014-07 which is strange as dvb.org says the 2019 is the initial version...? Whatever, it looks like I can now alter the colour scheme to highlight movies (in cyan) and sports (in green); the rest will be the alternating pattern as before, and the scheduling markers...
However, this too is not a big priority. Getting an EPG up on-screen (even if no programme details initially) is the priority.

PS: Oh, look, we're talking about RISC OS. ☺

Your comments:

Please note that while I check this page every so often, I am not able to control what users write; therefore I disclaim all liability for unpleasant and/or infringing and/or defamatory material. Undesired content will be removed as soon as it is noticed. By leaving a comment, you agree not to post material that is illegal or in bad taste, and you should be aware that the time and your IP address are both recorded, should it be necessary to find out who you are. Oh, and don't bother trying to inline HTML. I'm not that stupid! ☺
As of February 2025, commenting is no longer available to UK residents, following the implementation of the vague and overly broad Online Safety Act. You must tick the box below to verify that you are not a UK resident, and you expressly agree if you are in fact a UK resident that you will indemnify me (Richard Murray), as well as the person maintaining my site (Rob O'Donnell), the hosting providers, and so on. It's a shitty law, complain to your MP.
It's not that I don't want to hear from my British friends, it's because your country makes stupid laws.
 
You can now follow comment additions with the comment RSS feed. This is distinct from the b.log RSS feed, so you can subscribe to one or both as you wish.

jgh, 16th October 2025, 02:58

I remember having this problem parsing TV schedules on broadcast teletext. Just as you'd got the parsing working, the broadcasters would change the layout and you'd either have to change the parsing code or build yet more intelligence and exception trees into the parsing code without breaking it for pages that hadn't changed, or changed in different ways.    More recently, every time Richard Russell gets his Ceefax "emulator" working, fetching web data and presenting it as a teletext page, yep, the broadcasters mess with the data content.

Add a comment (v0.12) [help?] . . . try the comment feed!
		Your name
		Your email	(optional)
		Validation	Are you real? Please type 75381 backwards.
		UK resident	Check to verify that you are NOT a UK resident.
		Your comment