1 Update that simple tokenizer compulsory test so things will compile
5 (done) Fix #include <stdio.h> to read include directive correctly
7 txt/orig state of affairs:
9 The problem is that there are two ways to interpret line,col:
13 This isn't a problem when txt and orig point to the same character, as in:
18 index /*Here, the backslash break should be gobbled up by the space identifier*/
20 line,col has no ambiguity as to where it should point. However, when they point to different characters (i.e. at the beginning of a line):
25 line,col could either point to orig or to the first real character. Thus, we will do the latter.
27 Moreover, will a newline followed by backslash breaks generate a token that gobbles up said breaks? I believe it will, but no need to call this mandatory.
29 Thus, on a lookup with a txt pointer, the line/col/orig should match the real character and not preceding backslash breaks.
32 I've been assuming that every token starts with its first character, neglecting the case where a line starts with backslash breaks. The question is, given the txt pointer to the first character, where should the derived orig land?
34 Currently, the orig lands after the beginning backslash breaks, when instead it should probably land before them.
36 Here's what the tokenizer's text anchoring needs:
37 Broken/unbroken text pointer -> line/col
38 Unbroken contents per token to identify identifier text
39 Original contents per token to rebuild the document
40 Ability to change "original contents" so the document will be saved with modifications
41 Ability to insert new tokens
44 New tokens will typically have identical txt and orig, yea even the same pointer.
45 txt/txt_size for unbroken contents, orig/orig_size for original
46 modify orig to change the document
47 txt identifies identifier text
48 Line lookup tables are used to resolve txt/orig pointers; other pointers can't be resolved in the same fashion and may require traversing backward through the list.
51 Token txt/txt_size, orig/orig_size, orig_lines, txt_lines, and tok_point_lookup are all still correct.
52 Token line,col will be removed
54 Other improvements to do:
55 Sanity check the point lookups like crazy
56 Remove the array() structures in token_list, as these are supposed to be read-only
58 Make sure tok_point_lookup returns correct values for every single pointer possible, particularly those in orig that are on backslash-breaks
60 Convert the tok_message_queue into an array of messages bound to tokens.
62 Ask Rusty about the trailing newline in this case:
69 Here, rather than the trailing space being blank, it is "blank" from the comment perspective.
70 May require deeper analysis.
72 Todos from ccan_tokenizer.h
74 Assumption: Every token fits in one and exactly one line
76 Backslash-broken lines
79 Checks to implement in the tokenizer:
81 is the $ character used in an identifier (some configurations of GCC allow this)
82 are there potentially ambiguous sequences used in a string literal (e.g. "\0000")
83 Are there stray characters? (e.g. '\0', '@', '\b')
84 Are there trailing spaces at the end of lines (unless said spaces consume the entire line)?
85 Are there trailing spaces after a backslash-broken line?
90 backslash-newline sequence should register as an empty character, and the tokenizer's line value should be incremented accordingly.
93 Lex angle bracket strings in #include
95 Check the rules in the documentation
97 Examine the message queue as part of testing the tokenizer:
98 Make sure there are no bug messages
99 Make sure files compile with no warnings
100 For the tokenizer sanity check, make sure integers and floats have valid suffixes respectively
101 (e.g. no TOK_F for an integer, no TOK_ULL for a floating)
103 Update the scan_number sanity checks
104 (done) Move scan_number et al. to a separate C file
107 Overflow and underflow floats
109 (done) 0755f //octal 0755 with invalid suffix
110 (done) 0755e1 //floating 7550
112 Figure out how keywords will be handled.
113 Preprocessor directives are <strike>case-insensitive</strike> actually case-sensitive (except __VA_ARGS__)
114 All C keywords are case sensitive
115 __VA_ARGS__ should be read as an identifier unless it's in the expansion of a macro. Otherwise, GCC generates a warning.
116 We are in the expansion of a macro after <startline> <space> # <space>
117 Don't forget about __attribute__
118 Except for __VA_ARGS__, all preprocessor keywords are proceeded by <startline> <space> # <space>
121 All the words themselves will go into one opkw dictionary, and for both type and opkw, no distinction will be made between preprocessor and normal keywords.
122 Instead, int type will become short type; unsigned short cpp:1;
125 Commit ccan_tokenizer to the ccan repo
126 Introduce ccan_tokenizer to ccanlint
128 Write testcases for scanning all available operators
129 Support integer and floating point suffices (e.g. 500UL, 0.5f)
130 Examine the message queue after tokenizing
131 Make sure single-character operators have an opkw < 128
132 Make sure c_dictionary has no duplicate entries
133 Write verifiers for other types than TOK_WHITE
137 Operator table has been organized
138 Merged Rusty's changes
139 Fixed if -> while in finalize
140 Fixed a couple mistakes in run-simple-token.c testcases themselves
141 Expected orig/orig_size sizes weren't right
142 Made token_list_sanity_check a public function and used it throughout run-simple-token.c
143 Tests succeed and pass valgrind
145 Lines/columns of every token are recorded
147 (done) Fix "0\nstatic"
148 (done) Write tests to make sure backslash-broken lines have correct token locations.
149 (done) Correctly handle backslash-broken lines
150 One plan: Separate the scanning code from the reading code. Scanning sends valid ranges to reading, and reading fills valid tokens for the tokenizer/scanner to properly add
151 Another plan: Un-break backslash-broken lines into another copy of the input. Create an array of the positions of each real line break so
152 Annotate message queue messages with current token
156 Position in unbroken text
159 Real offset from start of line
161 Thus, we want an array of real line start locations wrt the unbroken text
167 <LINE>Here is a bro<LINE>ken line. Here is a
170 If we know the position of the token text wrt the unbroken text, we can look up the real line number and offset using only the array of real line start positions within the unbroken text.
172 Because all we need is the orig and orig_size with respect to the unbroken text to orient