6502/CLK
1
0
Fork 0
mirror of https://github.com/TomHarte/CLK.git synced 2024-11-19 23:32:28 +00:00
Code Issues Projects Releases Wiki Activity
3caa4705ca
CLK/Storage/Disk/PCMPatchedTrack.hpp

76 lines
2.2 KiB
C++
Raw Normal View History

This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
//
// PCMPatchedTrack.hpp
// Clock Signal
//
// Created by Thomas Harte on 15/12/2016.
// Copyright © 2016 Thomas Harte. All rights reserved.
//
#ifndef PCMPatchedTrack_hpp
#define PCMPatchedTrack_hpp
Introduced a test within the disk controller so as not to request illegal tracks from disks, instead automatically substituting an 'unformatted' track. Which is just empty.
2017-08-16 01:52:12 +00:00
#include "Disk.hpp"
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook `list` rather than `vector` stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the `Period`s aren't yet honoured.
2016-12-19 12:42:43 +00:00
#include "PCMSegment.hpp"
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
namespace Storage {
namespace Disk {
/*!
A subclass of @c Track that patches an existing track with PCM segments.
*/
class PCMPatchedTrack: public Track {
public:
/*!
Constructs a @c PCMPatchedTrack that will return events from @c underlying_track in
regions where it has not had alternative PCM data installed.
*/
Ensured the patchable track owns its underlying track.
2016-12-17 23:17:22 +00:00
PCMPatchedTrack(std::shared_ptr<Track> underlying_track);
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
After a quick bit of reading, discovered the virtual copy constructor pattern really is only a convention in C++, and conformed to it. Which hopefully gives copyable tracks.
2016-12-30 22:25:39 +00:00
/*!
Copy constructor, for Track.
*/
PCMPatchedTrack(const PCMPatchedTrack &);
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
/*!
Replaces whatever is currently on the track from @c start_position to @c start_position + segment length
with the contents of @c segment.
Added an easy way for disk controllers to clamp termination of written data exactly to the index hole. This commit also temporarily provides a whole load of extra logging and minor logic improvements from the 8272. I'm mid-flow on finding a particularly vicious error in its handling of writing; wait for the pull request. But, at least: now waits for the first part of a post-ID gap before writing data, and attempts partially to handle appearance of the index hole during writing a track. More work to do on that though.
2017-08-15 20:05:10 +00:00
@param start_time The time at which this segment begins. Must be in the range [0, 1).
@param segment The PCM segment to add.
@param clamp_to_index_hole If @c true then the new segment will be truncated if it overruns the index hole;
it will otherwise write over the index hole and continue.
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
*/
Added an easy way for disk controllers to clamp termination of written data exactly to the index hole. This commit also temporarily provides a whole load of extra logging and minor logic improvements from the 8272. I'm mid-flow on finding a particularly vicious error in its handling of writing; wait for the pull request. But, at least: now waits for the first part of a post-ID gap before writing data, and attempts partially to handle appearance of the index hole during writing a track. More work to do on that though.
2017-08-15 20:05:10 +00:00
void add_segment(const Time &start_time, const PCMSegment &segment, bool clamp_to_index_hole);
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
// To satisfy Storage::Disk::Track
Event get_next_event();
Opted to pass times by reference and added enough to PCMPatchedTrack that it could start being used by the disk controller, albeit that it doesn't work.
2016-12-17 21:26:45 +00:00
Time seek_to(const Time &time_since_index_hole);
After a quick bit of reading, discovered the virtual copy constructor pattern really is only a convention in C++, and conformed to it. Which hopefully gives copyable tracks.
2016-12-30 22:25:39 +00:00
Track *clone();
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
private:
Ensured the patchable track owns its underlying track.
2016-12-17 23:17:22 +00:00
std::shared_ptr<Track> underlying_track_;
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook `list` rather than `vector` stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the `Period`s aren't yet honoured.
2016-12-19 12:42:43 +00:00
struct Period {
Time start_time, end_time;
Time segment_start_time;
Switched to much more logical `shared_ptr` ownership of `PCMSegmentEventSource`s by `Period`s.
2016-12-20 23:13:10 +00:00
std::shared_ptr<PCMSegmentEventSource> event_source; // nullptr => use the underlying track
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook `list` rather than `vector` stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the `Period`s aren't yet honoured.
2016-12-19 12:42:43 +00:00
Attempted properly to cover the exactly-equal starts and ends cases, and to improve meaning.
2016-12-20 23:32:49 +00:00
void push_start_to_time(const Storage::Time &new_start_time);
void trim_end_to_time(const Storage::Time &new_end_time);
Switched to much more logical `shared_ptr` ownership of `PCMSegmentEventSource`s by `Period`s.
2016-12-20 23:13:10 +00:00
Period(const Time &start_time, const Time &end_time, const Time &segment_start_time, std::shared_ptr<PCMSegmentEventSource> event_source) :
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook `list` rather than `vector` stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the `Period`s aren't yet honoured.
2016-12-19 12:42:43 +00:00
start_time(start_time), end_time(end_time), segment_start_time(segment_start_time), event_source(event_source) {}
`Period`s need a custom copy constructor too, if they're going to avoid sharing an event_source.
2016-12-30 22:39:52 +00:00
Period(const Period &);
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
};
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook `list` rather than `vector` stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the `Period`s aren't yet honoured.
2016-12-19 12:42:43 +00:00
std::vector<Period> periods_;
Made an attempt fully to implement PCMPatchedTrack. Which now requires tests.
2016-12-20 12:30:57 +00:00
std::vector<Period>::iterator active_period_;
With lots of logging arising temporarily, fixed bug whereby conversion to a patched track would lead to holding a track with a distinct measure of time, leading to improperly-placed patches.
2016-12-26 03:00:39 +00:00
Time current_time_, insertion_error_;
Made an attempt to come up with a data structure that actually makes sense (though perhaps this is textbook `list` rather than `vector` stuff? I guess it depends on the frequency I expect inserts to occur versus reads) and to implement inserts. Though the `Period`s aren't yet honoured.
2016-12-19 12:42:43 +00:00
void insert_period(const Period &period);
This is going to be a slow walk, I think. This class attempts to be the scratchpad which will hold in-memory track modifications.
2016-12-17 00:20:38 +00:00
};
}
}
#endif /* PCMPatchedTrack_hpp */
Reference in New Issue Copy Permalink