haskal
/
kicad
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
kicad/pcbnew/plugins/altium/altium_parser.ksy

753 lines
14 KiB
Plaintext
Raw Normal View History

Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
# Can be viewed in: https://ide.kaitai.io/
#
# This file is a formal specification of the binary format used in Altium.
# Files need to manually extracted using a program which can read the Microsoft Compound File Format.
#
# While I do not create a parser using this file, it is still very helpful to understand the binary
# format.
meta:
id: altium_binary
endian: le
encoding: ISO8859-1
seq:
altium: extend documentation so we can parse Altium *.PcbLib files in the future
2022-01-05 19:07:42 +00:00
# Only relevant for *.PcbLib files
# - id: name
# type: name
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
- id: record
type: record
repeat: eos
# https://github.com/thesourcerer8/altium2kicad/blob/master/convertpcb.pl#L1291
types:
altium: extend documentation so we can parse Altium *.PcbLib files in the future
2022-01-05 19:07:42 +00:00
name:
seq:
- id: sub1_len
type: u4
- id: data
type: name_sub1
size: sub1_len
name_sub1:
seq:
- id: name_len
type: u1
- id: name
size: name_len
type: str
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
record:
seq:
- id: recordtype
type: u1
enum: record_id
- id: record
type:
switch-on: recordtype
cases:
record_id::arc6: arc
record_id::pad6: pad
record_id::via6: via
record_id::track6: track
record_id::text6: text
record_id::fill6: fill
record_id::region6: region
altium: initial code to import STEP models For now, they are extracted into the same directory as the project file. Furthermore, there seems to be a problem with rotation / positioning of some models.
2020-04-21 13:34:06 +00:00
record_id::componentbody6: componentbody
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
arc:
seq:
- id: sub1_len
type: u4
- id: data
type: arc_sub1
size: sub1_len
arc_sub1:
seq:
- id: layer
type: u1
- #id: flags_u7
type: b1
- #id: flags_u6
type: b1
- #id: flags_u5
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- id: is_not_polygonoutline
type: b1
- #id: flags_u0
type: b1
- id: is_keepout
type: u1 # KEEPOUT = 2
- id: net
type: u2
- id: subpolyindex
type: u2
- id: component
type: u2
- size: 4
- id: center
type: xy
- id: radius
type: u4
- id: start_angle
type: f8
- id: end_angle
type: f8
- id: width
type: u4
pad:
seq:
- id: sub1_len
type: u4
- id: designator
type: pad_sub1
size: sub1_len
- id: sub2_len
type: u4
- size: sub2_len
- id: sub3_len
type: u4
- size: sub3_len
- id: sub4_len
type: u4
- size: sub4_len
- id: sub5_len
type: u4
- id: size_and_shape
type: pad_sub5
size: sub5_len
- id: sub6_len
type: u4
- id: size_and_shape_by_layer
type: pad_sub6
size: sub6_len
if: sub6_len > 0
pad_sub1:
seq:
- id: name_len # = len-1?
type: u1
- id: name
type: str
size: name_len
pad_sub5:
seq:
- id: layer # $pos+23
type: u1
enum: layer
- id: test_fab_top
type: b1
- id: tent_bottom
type: b1
- id: tent_top
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- #id: flags_u1
type: b1
- #id: flags_u0
type: b1
- #id: flags2_u7
type: b1
- #id: flags2_u6
type: b1
- #id: flags2_u5
type: b1
- #id: flags2_u4
type: b1
- #id: flags2_u3
type: b1
- #id: flags2_u2
type: b1
- #id: flags2_u1
type: b1
- id: test_fab_bottom
type: b1
#- id: u
# size: 1
- id: net # $pos+26
type: u2
- size: 2
- id: component # $pos+30
type: u2
- size: 4
- id: position # $pos+36, $pos+40
type: xy
- id: topsize # $pos+44, $pos+48
type: xy
- id: midsize # $pos+52, $pos+56
type: xy
- id: botsize # $pos+60, $pos+64
type: xy
- id: holesize # $pos+68
type: u4
- id: topshape # $pos+72
type: u1
enum: pad_shape
- id: midshape # $pos+73
type: u1
enum: pad_shape
- id: botshape # $pos+74
type: u1
enum: pad_shape
- id: direction # $pos+75
type: f8
- id: plated # $pos+83
type: u1
enum: boolean
- size: 1
- id: pad_mode # $pos+85
type: u1
enum: pad_mode
- size: 5
- id: ccw # $pos+91
type: u4
- id: cen # $pos+95
type: u1
- size: 1
- id: cag # $pos+97
type: u4
- id: cpr # $pos+101
type: u4
- id: cpc # $pos+105
type: u4
- id: pastemaskexpanionmanual
type: s4
- id: soldermaskexpansionmanual # $pos+113
type: s4
- id: cpl # $pos+117
type: u1
- size: 6
- id: pastemaskexpansionmode # $pos+124
type: u1
enum: pad_mode_rule
- id: soldermaskexpansionmode # $pos+125
type: u1
enum: pad_mode_rule
- size: 3
- id: holerotation # $pos+129
type: f8
- size: 4
- id: testpoint_assembly_top
type: u1
enum: boolean
- id: testpoint_assembly_bottom
type: u1
enum: boolean
pad_sub6:
seq:
- id: x
type: s4
repeat: expr
repeat-expr: 29
- id: y
type: s4
repeat: expr
repeat-expr: 29
- id: shape
type: u1
enum: pad_shape
repeat: expr
repeat-expr: 29
- size: 1
- id: hole_type
type: u1
enum: pad_hole_type
- id: slot_length
type: s4
- id: slot_rotation
type: f8
- id: holeoffset_x
type: s4
repeat: expr
repeat-expr: 32
- id: holeoffset_y
type: s4
repeat: expr
repeat-expr: 32
- size: 1
- id: shape_alt
type: u1
enum: pad_shape_alt
repeat: expr
repeat-expr: 32
- id: corner_radius
type: u1
repeat: expr
repeat-expr: 32
- size: 32
via:
seq:
- id: sub1_len
type: u4
- id: data
type: via_sub1
size: sub1_len
via_sub1:
seq:
- size: 1
- id: test_fab_top
type: b1
- id: tent_bottom
type: b1
- id: tent_top
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- #id: flags_u1
type: b1
- #id: flags_u0
type: b1
- #id: flags2_u7
type: b1
- #id: flags2_u6
type: b1
- #id: flags2_u5
type: b1
- #id: flags2_u4
type: b1
- #id: flags2_u3
type: b1
- #id: flags2_u2
type: b1
- #id: flags2_u1
type: b1
- id: test_fab_bottom
type: b1
- id: net
type: u2
- size: 2
- id: component
type: u2
- size: 4
- id: pos # 13
type: xy
- id: diameter # 21
type: s4
- id: holesize # 29
type: s4
- id: start_layer
type: u1
enum: layer
- id: end_layer
type: u1
enum: layer
- size: 43
- id: via_mode
type: u1
enum: pad_mode
- id: diameter_alt
type: s4
repeat: expr
repeat-expr: 32
track:
seq:
- id: sub1_len
type: u4
- id: data
type: track_sub1
size: sub1_len
track_sub1:
seq:
- id: layer
type: u1
enum: layer
- #id: flags_u7
type: b1
- #id: flags_u6
type: b1
- #id: flags_u5
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- id: is_not_polygonoutline
type: b1
- #id: flags_u0
type: b1
- id: is_keepout
type: u1 # KEEPOUT = 2
- id: net
type: u2
- id: subpolyindex
type: u2
- id: component
type: u2
- size: 4
- id: start # 13
type: xy
- id: end # 21
type: xy
- id: width # 29
type: s4
text:
seq:
- id: sub1_len
type: u4
- id: properties
type: text_sub1
size: sub1_len
- id: sub2_len
type: u4
- id: text
type: text_sub2
size: sub2_len
text_sub1:
seq:
- id: layer
type: u1
- #id: flags_u7
type: b1
- #id: flags_u6
type: b1
- #id: flags_u5
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- #id: flags_u1
type: b1
- #id: flags_u0
type: b1
- size: 1
- id: net
type: u2
- size: 2
- id: component
type: u2
- size: 4
- id: pos
type: xy
- id: height
type: u4
- id: font_name_id
type: u1
- size: 1
- id: rotation
type: f8
altium: parse text properties (bold, italic, inverted)
2020-05-10 16:51:38 +00:00
- id: is_mirrored
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
type: u1
enum: boolean
- id: strokewidth
type: u4
- id: is_comment
type: u1
enum: boolean
- id: is_designator
type: u1
enum: boolean
altium: parse text properties (bold, italic, inverted)
2020-05-10 16:51:38 +00:00
- size: 2
- id: is_bold
type: u1
enum: boolean
- id: is_italic
type: u1
enum: boolean
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
- id: font_name
size: 64
type: str # TODO: terminates with [0, 0]
encoding: UTF-16
altium: parse text properties (bold, italic, inverted)
2020-05-10 16:51:38 +00:00
- id: is_inverted
type: u1
enum: boolean
altium: detect barcode objects, improve binary documentation for text objects
2020-05-10 18:11:24 +00:00
- id: margin
type: s4
- id: use_offset # use margin otherwise
type: u1
enum: boolean
- size: 16
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
- id: position
type: u1
enum: text_position
altium: detect barcode objects, improve binary documentation for text objects
2020-05-10 18:11:24 +00:00
- id: offset
type: s4
- id: barcode_full_size # TODO: also for non-barcode?
type: xy
- id: barcode_margin # TODO: also for non-barcode?
type: xy
- size: 4
- id: barcode_type
type: u1
enum: text_barcode_type
- size: 1
- id: barcode_inverted
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
type: u1
enum: boolean
altium: detect barcode objects, improve binary documentation for text objects
2020-05-10 18:11:24 +00:00
- id: font_type
type: u1
enum: text_font_type
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
- id: barcode_name
size: 64
type: str # TODO: terminates with [0, 0]
encoding: UTF-16
altium: detect barcode objects, improve binary documentation for text objects
2020-05-10 18:11:24 +00:00
- id: barcode_show_text
type: u1
enum: boolean
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
text_sub2:
seq:
- id: len
type: u1
- id: name
type: str
size: len
fill:
seq:
- id: sub1_len
type: u4
- id: data
type: fill_sub1
size: sub1_len
fill_sub1:
seq:
- id: layer
type: u1
- #id: flags_u7
type: b1
- #id: flags_u6
type: b1
- #id: flags_u5
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- #id: flags_u1
type: b1
- #id: flags_u0
type: b1
- id: is_keepout
type: u1 # KEEPOUT = 2
- id: net
type: u2
- size: 2
- id: component
type: u2
- size: 4
- id: pos1
type: xy
- id: pos2
type: xy
- id: rotation
type: f8
region:
seq:
- id: sub1_len
type: u4
- id: data
type: region_sub1
size: sub1_len
region_sub1:
seq:
- id: layer
type: u1
- #id: flags_u7
type: b1
- #id: flags_u6
type: b1
- #id: flags_u5
type: b1
- #id: flags_u4
type: b1
- #id: flags_u3
type: b1
- id: is_not_locked
type: b1
- #id: flags_u1
type: b1
- #id: flags_u0
type: b1
- id: is_keepout
type: u1 # KEEPOUT = 2
- id: net
type: u2
altium: parse Regions6 to fill zone on import
2021-02-06 22:06:23 +00:00
- id: subpolyindex
type: u2
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
- id: component
type: u2
- size: 5
- id: holecount # TODO: check
type: u2
- size: 2
- id: propterties_len
type: u4
- id: properties
size: propterties_len
type: str
altium: parse Regions6 to fill zone on import
2021-02-06 22:06:23 +00:00
# region1 type
- id: outline
type: region1_part
- id: holes
type: region1_part
repeat-expr: holecount
repeat: expr
# TODO
#- id: vertices_num
# type: u4
#- id: vertices2 # region2 type
# repeat: expr
# repeat-expr: vertices_num+1
# type: xyf2
region1_part:
seq:
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
- id: vertices_num
type: u4
- id: vertices # region1 type
repeat: expr
repeat-expr: vertices_num
type: xyf
altium: initial code to import STEP models For now, they are extracted into the same directory as the project file. Furthermore, there seems to be a problem with rotation / positioning of some models.
2020-04-21 13:34:06 +00:00
componentbody:
seq:
- id: sub1_len
type: u4
- id: data
type: componentbody_sub1
size: sub1_len
componentbody_sub1:
seq:
- size: 7
- id: component
type: u2
- size: 9
- id: propterties_len
type: u4
- id: properties
size: propterties_len
type: str
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
xy:
seq:
- id: x
type: s4
- id: y
type: s4
xyf: # no idea why two different formats?
seq:
- id: x
type: f8
- id: y
type: f8
xyf2: # no idea why two different formats?
seq:
- id: is_round
type: u1
enum: boolean
- id: position
type: xy
- id: center
type: xy
- id: radius
type: u4
- id: angle1
type: f8
- id: angle2
type: f8
enums:
record_id:
0x01: arc6
0x02: pad6
0x03: via6
0x04: track6
0x05: text6
0x06: fill6
0x0b: region6
altium: initial code to import STEP models For now, they are extracted into the same directory as the project file. Furthermore, there seems to be a problem with rotation / positioning of some models.
2020-04-21 13:34:06 +00:00
0x0c: componentbody6
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
boolean:
0: false
1: true
pad_shape:
0: unknown
1: circle
2: rect
3: octagonal
pad_shape_alt:
0: unknown
1: round
2: rect
3: octagonal
9: roundrectangle
pad_hole_type:
0: normal
1: square
2: slot
pad_mode:
0: simple
1: top_middle_bottom
2: full_stack
pad_mode_rule:
0: unknown
1: rule
2: manual
text_position:
1: left_top
2: left_center
3: left_bottom
4: center_top
5: center_center
6: center_bottom
7: right_top
8: right_center
9: right_bottom
altium: detect barcode objects, improve binary documentation for text objects
2020-05-10 18:11:24 +00:00
text_font_type:
0: stroke
1: truetype
2: barcode
text_barcode_type:
0: code39
1: code128
Use Record handling for parsing Now, we know how big a record is, and should be able to parse all boards without missing bytes?
2020-04-03 23:22:24 +00:00
layer:
1: f_cu
32: b_cu