Partition Manager
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ... 29
Colin (478) 2433 posts |
Just to clarify I didn’t mean to replace the utility with a spreadsheet like display just the list at the bottom. The reason I suggested it is that in the mockup the titles in the list look to me disjoint from the list and the columns blur into each other. I think a spreadsheet like arrangement could work. My suggestions were just a case of if I don’t ask I definitely don’t get. |
David Pitt (3386) 1248 posts |
With 0.21 :- The PartMan partitioned pen shows a number of “extra” partitions. The ‘Capacity’ values are still wrong. For example Free for SCSI::0, Pen16, shows Free 8GB, Used 6GB and Size 14GB. |
Steve Pampling (1551) 8125 posts |
Opinions, we all have them. Fortunately I don’t think we’re likely to start a riot :) |
Doug Webb (190) 1156 posts |
0.21 works again but still no SDFS::1 Also SCSI::4, which is my main SSD, 224GB, says SCSI::4/Filecore 223.57GB/ SSD/Filecore and 3.57GB in the pictoral section and in the text section Layout SSD (SCSI::4)- E, New Map, Format Filecore, Capacity 3.57 with free 7.54GB. Others are similar. Also if a window goes over the PartManager window it deletes part of the display as does moving/resizing the window so the redraw code may need tweaking. I can’t put a picture as I have no webspace to place and point to via Textile unless someone else knows how to insert a picture directly? |
Andrew McCarthy (3688) 602 posts |
Thanks for v0.21, I get the following message Parameters not recognised in “add_drives_SCSIFS” at line 87 |
Jeff Doggett (257) 233 posts |
Don’t be fooled by the presence of *_DiscOp64. |
Doug Webb (190) 1156 posts |
Just thought could this being tripped up by how the confuguration of drives is set up on the ARMX6 by RComp. Essentialy it has a built in card reader and also the SDFS drives are set up with some obey files that RComp include. So I have: I also have three ShareFS drive icons for remote systems so that brings up the total of eight drives? SDFS::1 details from RComp: WARNING DO NOT FORMAT this disc with !HForm. The OS is housed in 8Mb of “invisible” So does the partiticular partition set up of this card cause PartMgr not to report it? HideSDFS obey in Boot.Tasks: REM OS_UpCall 24 to show icon, 25 to hide OpenHD4 obey file in Boot.Tasks BootDir$=FNGetSysVar(“Boot$dir”) END |
Colin (478) 2433 posts |
The ArmX6 boots from SDFS:1 and the bootable sd card stores its boot files in an area of the disc marked as defective. According to DiscKnight the defect list starts at 0xA000. |
Doug Webb (190) 1156 posts |
Thanks Colin as I said I wondered if the way it was partitioned, or in this case not partitioned but defected out, might cause an issue with Jon’s attempts to fully read the SDFS::1 SDCard. As they say “Not all partitioned drives are the same or equal” :-) |
Colin (478) 2433 posts |
Other programs – hform, DiscKnight, SystemDisc – have no problem finding the drive so its odd that PartMan is having problems unless it is being rejected as unreadable. I wonder if a disc is displayed if the format is unreadable. |
Alan Adams (2486) 1140 posts |
This is BASIC, not Obey. However my surprise is the second line – is it really LEN? (and why is Textile underlining the LEN?) OK, I think the question is caused by Textile. Typing LEN followed immediately by (BootDir$) comes out as LEN with an underline. I had to add a space to get it to show what I meant in the question. (and it’s taken 8 edits to get this to display as I intended. Textile? Bah, humbug.) |
Rick Murray (539) 13750 posts |
Because something in capitals followed by something in brackets is interpreted as an acronym with a following explanation that can appear as a pop-up (on capable browsers).
You mean LEN(BootDir$)? ;-) There are two cheats that may be employed here. The first is to wrap the content in The second is to wrap the text in ‘@’ symbols which will force it to be Note, by the way, that detected URLs will be turned into links, even if this is wrong and breaks stuff. Try an http link in the comments of some code, you’ll see the mess it makes as it converts the link into an actual link, and then embeds the actual link markup in the comment. Hell, here’s an example:
What you see… is not what I wrote.
Yeah, Textile is… difficult. There’s apparently a full detailed spec of exactly how Textile behaves with various sorts of markup. Maybe one day Textile itself will bother to read that spec. ;-) |
Doug Webb (190) 1156 posts |
Yep sorry about that. “Keep up at the back” and as they also say “should of gone to Specsavers”, other opticians are available, as looking again at the files it is clear they are Basic files.
Just there to catch you out when you can’t even put something simple in to the comments. |
Steve Fryatt (216) 2095 posts |
Doesn’t marking it as code (like this That’s written as
by the way. If you’re writing a block of code, prefix the first line with |
Steve Pampling (1551) 8125 posts |
Indeed. As Rick said:
However, he also said speaking in jest but catching truth:
Anyone who says they’ve never had textile do something “different” is either telling porkies or needs to play the lottery, so we can get proper RO development funding. |
Jon Abbott (1421) 2640 posts |
v0.22 is up with some minor bug fixes. I’m not going have much time to spend on it during the week, so it will be the weekend before I can sort out debug output to diagnose some of the issues reported above. With regard to the window not redrawing correctly – I’m guessing I’ve done something wrong in the icons, that is causing the Wimp to fail to redraw them correctly. |
David J. Ruck (33) 1585 posts |
I couldn’t remember how I looked for drives to put in the DisKnight front end menus, but I found a bit of old BASIC test code. REM >EnumDrives ON ERROR PRINTREPORT$;" at line ";ERL:END DIM buff% 256 M=0:I=0 REPEAT OM=M:OI=I SYS"XOS_Module",12,M,I TO R0,M,I,B,W,P IFR0<>12 END B=B+B!&10 M$="":REPEAT:M$+=CHR$?B:B+=1:UNTIL?B<32 P$="":REPEAT:P$+=CHR$?P:P+=1:UNTIL?P<32 IFM$="FileCore" ANDP$<>"Base" THEN FORD%=0 TO 7 path$=P$+"::"+STR$D%+".$" SYS"XOS_FSControl",37,path$,buff%,0,0,256 TO ,,,,,R buff%?(256-R)=13 IF$buff%<>"" PRINTP$;" ";D%;" ";$buff% NEXT ENDIF UNTIL0
God that’s ancient, it was for D-Format discs which needed manual compaction to stop the free space table getting full after just 82 entries. It did have some mild warming over to recognise E and E+ drives, but it doesn’t know about any more recent disc types such as SD cards, and fails completely on some FS’s. Comment out the line in !Boot which makes it take over from the standard Free Space, without running it first. |
Jon Abbott (1421) 2640 posts |
I’ve yet to figure out how to correctly distinguish between old/new map or D, E, E+, F?, F+? layout ADFS drives as documentation seems to be thin on the ground. Is there a known method to save reinventing the wheel? |
David J. Ruck (33) 1585 posts |
FreeMap cheated by parsing the output of the *Map command looking for big map for E format, and big directories for E+ format. It wasn’t until I wrote DisKnight I had to get my hands grubby by looking at the data on the disc. The good news is it is all in the PRMs for E format, and E+ was in the Ursula documentation (which may still be on the ROL site), the bad news is it is a real head fsck to understand. DiscKnight doesn’t know anything about a D format drive, and would just get very upset if you presented it with one. E and E+ drives have the same disc record and map structures, but their directories differ, but are clearly marked by differing first and last words. It’s not cached in my head at the moment, but I could probably give some more pointers if pressed. Oh and F and F+ formats are the same as E and E+, the E and F are use to distinguish between 800K and 1.6M floppies. The conversion used by Acorn was to call the format for harddiscs E/E+, although lots of people incorrectly described them as F/F+. |
David Pitt (3386) 1248 posts |
PartMgr 0.22 looks good here, now showing capacities as expected. Many thanks. |
Doug Webb (190) 1156 posts |
+1 Though SDFS::1 still missing in action. FreeMap disabled though I did like, to me at least, it’s better graphical display and thats all it was used for. |
Andrew McCarthy (3688) 602 posts |
After a bit of unplugging, v0.22 works if I unplug my USB SSD which worked with version v0.15 PiBoot (SDFS::0.$.?) If I don’t unplug the attached SSD drive, then this version errors with Parameters not recognised in “add_drives_SCSIFS” at line 87" |
Martin Avison (27) 1479 posts |
Please could SATA connected ADFS drives be included when you get chance? |
Jon Abbott (1421) 2640 posts |
Thanks for the info, I suspect my switch from internally using addresses to LBA’s has broken it.
ADFS is next on my list
I noted the disc record is the same for E/E+. I’ve assumed that if *_DescribeDisc returns with big_flag in a large record it’s E(+) and if it returns a small record its D. I’m not sure I need worry about directory type as it’s only the format we’re concerned with in a partition manager.
I did wonder if they were floppy only. I also noted the idiotic ++ convention, with I think is a RO4/6 format. Do I distinguish between E/E+ by the size of idlen? |
David J. Ruck (33) 1585 posts |
Sorry that was a typo from me. ROL did mess with the disc format by adding things such as the display and sort order to directories, but they refused to give me any information, so if DiscKnight see’s any reserved bits being non-zero it will zero them until such time as they are formally documented.
No, its the by FileCore version at offset 44 of the disc record, should be 0 for E and 1 for E+. Those pages really need updating for E+ format. |
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 ... 29