diff options
Diffstat (limited to 'lib/pure/parsecfg.nim')
| -rw-r--r-- | lib/pure/parsecfg.nim | 337 |
1 files changed, 202 insertions, 135 deletions
diff --git a/lib/pure/parsecfg.nim b/lib/pure/parsecfg.nim index c8e936ef1..8a43daf54 100644 --- a/lib/pure/parsecfg.nim +++ b/lib/pure/parsecfg.nim @@ -7,8 +7,8 @@ # distribution, for details about the copyright. # -## The ``parsecfg`` module implements a high performance configuration file -## parser. The configuration file's syntax is similar to the Windows ``.ini`` +## The `parsecfg` module implements a high performance configuration file +## parser. The configuration file's syntax is similar to the Windows `.ini` ## format, but much more powerful, as it is not a line based parser. String ## literals, raw string literals and triple quoted string literals are supported ## as in the Nim programming language. @@ -19,106 +19,173 @@ ## :literal: ## ## Here is an example of how to use the configuration file parser: -## -## .. code-block:: nim -## -## import os, parsecfg, strutils, streams -## -## var f = newFileStream(paramStr(1), fmRead) -## if f != nil: -## var p: CfgParser -## open(p, f, paramStr(1)) -## while true: -## var e = next(p) -## case e.kind -## of cfgEof: break -## of cfgSectionStart: ## a ``[section]`` has been parsed -## echo("new section: " & e.section) -## of cfgKeyValuePair: -## echo("key-value-pair: " & e.key & ": " & e.value) -## of cfgOption: -## echo("command: " & e.key & ": " & e.value) -## of cfgError: -## echo(e.msg) -## close(p) -## else: -## echo("cannot open: " & paramStr(1)) -## -## -## Examples -## ======== -## +runnableExamples("-r:off"): + import std/[strutils, streams] + + let configFile = "example.ini" + var f = newFileStream(configFile, fmRead) + assert f != nil, "cannot open " & configFile + var p: CfgParser + open(p, f, configFile) + while true: + var e = next(p) + case e.kind + of cfgEof: break + of cfgSectionStart: ## a `[section]` has been parsed + echo "new section: " & e.section + of cfgKeyValuePair: + echo "key-value-pair: " & e.key & ": " & e.value + of cfgOption: + echo "command: " & e.key & ": " & e.value + of cfgError: + echo e.msg + close(p) + +##[ ## Configuration file example -## -------------------------- -## -## .. code-block:: nim -## +]## + +## ```none ## charset = "utf-8" ## [Package] ## name = "hello" ## --threads:on ## [Author] -## name = "lihf8515" -## qq = "10214028" -## email = "lihaifeng@wxm.com" -## +## name = "nim-lang" +## website = "nim-lang.org" +## ``` + +##[ ## Creating a configuration file -## ----------------------------- -## .. code-block:: nim -## -## import parsecfg -## var dict=newConfig() -## dict.setSectionKey("","charset","utf-8") -## dict.setSectionKey("Package","name","hello") -## dict.setSectionKey("Package","--threads","on") -## dict.setSectionKey("Author","name","lihf8515") -## dict.setSectionKey("Author","qq","10214028") -## dict.setSectionKey("Author","email","lihaifeng@wxm.com") -## dict.writeConfig("config.ini") -## +]## + +runnableExamples: + var dict = newConfig() + dict.setSectionKey("","charset", "utf-8") + dict.setSectionKey("Package", "name", "hello") + dict.setSectionKey("Package", "--threads", "on") + dict.setSectionKey("Author", "name", "nim-lang") + dict.setSectionKey("Author", "website", "nim-lang.org") + assert $dict == """ +charset=utf-8 +[Package] +name=hello +--threads:on +[Author] +name=nim-lang +website=nim-lang.org +""" + +##[ ## Reading a configuration file -## ---------------------------- -## .. code-block:: nim -## -## import parsecfg -## var dict = loadConfig("config.ini") -## var charset = dict.getSectionValue("","charset") -## var threads = dict.getSectionValue("Package","--threads") -## var pname = dict.getSectionValue("Package","name") -## var name = dict.getSectionValue("Author","name") -## var qq = dict.getSectionValue("Author","qq") -## var email = dict.getSectionValue("Author","email") -## echo pname & "\n" & name & "\n" & qq & "\n" & email -## +]## + +runnableExamples("-r:off"): + let dict = loadConfig("config.ini") + let charset = dict.getSectionValue("","charset") + let threads = dict.getSectionValue("Package","--threads") + let pname = dict.getSectionValue("Package","name") + let name = dict.getSectionValue("Author","name") + let website = dict.getSectionValue("Author","website") + echo pname & "\n" & name & "\n" & website + +##[ ## Modifying a configuration file -## ------------------------------ -## .. code-block:: nim -## -## import parsecfg -## var dict = loadConfig("config.ini") -## dict.setSectionKey("Author","name","lhf") -## dict.writeConfig("config.ini") -## -## Deleting a section key in a configuration file -## ---------------------------------------------- -## .. code-block:: nim -## -## import parsecfg -## var dict = loadConfig("config.ini") -## dict.delSectionKey("Author","email") -## dict.writeConfig("config.ini") +]## + +runnableExamples("-r:off"): + var dict = loadConfig("config.ini") + dict.setSectionKey("Author", "name", "nim-lang") + dict.writeConfig("config.ini") -import - strutils, lexbase, streams, tables +##[ +## Deleting a section key in a configuration file +]## + +runnableExamples("-r:off"): + var dict = loadConfig("config.ini") + dict.delSectionKey("Author", "website") + dict.writeConfig("config.ini") + +##[ +## Supported INI File structure +]## + +# taken from https://docs.python.org/3/library/configparser.html#supported-ini-file-structure +runnableExamples: + import std/streams + + var dict = loadConfig(newStringStream("""[Simple Values] + key=value + spaces in keys=allowed + spaces in values=allowed as well + spaces around the delimiter = obviously + you can also use : to delimit keys from values + [All Values Are Strings] + values like this: 19990429 + or this: 3.14159265359 + are they treated as numbers : no + integers floats and booleans are held as: strings + can use the API to get converted values directly: true + [No Values] + key_without_value + # empty string value is not allowed = + [ Seletion A ] + space around section name will be ignored + [You can use comments] + # like this + ; or this + # By default only in an empty line. + # Inline comments can be harmful because they prevent users + # from using the delimiting characters as parts of values. + # That being said, this can be customized. + [Sections Can Be Indented] + can_values_be_as_well = True + does_that_mean_anything_special = False + purpose = formatting for readability + # Did I mention we can indent comments, too? + """) + ) + + let section1 = "Simple Values" + assert dict.getSectionValue(section1, "key") == "value" + assert dict.getSectionValue(section1, "spaces in keys") == "allowed" + assert dict.getSectionValue(section1, "spaces in values") == "allowed as well" + assert dict.getSectionValue(section1, "spaces around the delimiter") == "obviously" + assert dict.getSectionValue(section1, "you can also use") == "to delimit keys from values" + + let section2 = "All Values Are Strings" + assert dict.getSectionValue(section2, "values like this") == "19990429" + assert dict.getSectionValue(section2, "or this") == "3.14159265359" + assert dict.getSectionValue(section2, "are they treated as numbers") == "no" + assert dict.getSectionValue(section2, "integers floats and booleans are held as") == "strings" + assert dict.getSectionValue(section2, "can use the API to get converted values directly") == "true" + + let section3 = "Seletion A" + assert dict.getSectionValue(section3, + "space around section name will be ignored", "not an empty value") == "" + + let section4 = "Sections Can Be Indented" + assert dict.getSectionValue(section4, "can_values_be_as_well") == "True" + assert dict.getSectionValue(section4, "does_that_mean_anything_special") == "False" + assert dict.getSectionValue(section4, "purpose") == "formatting for readability" + +import std/[strutils, lexbase, streams, tables] +import std/private/decode_helpers +import std/private/since + +when defined(nimPreviewSlimSystem): + import std/syncio include "system/inclrtl" + type CfgEventKind* = enum ## enumeration of all events that may occur when parsing cfgEof, ## end of file reached - cfgSectionStart, ## a ``[section]`` has been parsed - cfgKeyValuePair, ## a ``key=value`` pair has been detected - cfgOption, ## a ``--key=value`` command line option + cfgSectionStart, ## a `[section]` has been parsed + cfgKeyValuePair, ## a `key=value` pair has been detected + cfgOption, ## a `--key=value` command line option cfgError ## an error occurred during parsing CfgEvent* = object of RootObj ## describes a parsing event @@ -126,12 +193,12 @@ type of cfgEof: nil of cfgSectionStart: section*: string ## `section` contains the name of the - ## parsed section start (syntax: ``[section]``) + ## parsed section start (syntax: `[section]`) of cfgKeyValuePair, cfgOption: key*, value*: string ## contains the (key, value) pair if an option - ## of the form ``--key: value`` or an ordinary - ## ``key= value`` pair has been parsed. - ## ``value==""`` if it was not specified in the + ## of the form `--key: value` or an ordinary + ## `key= value` pair has been parsed. + ## `value==""` if it was not specified in the ## configuration file. of cfgError: ## the parser encountered an error: `msg` msg*: string ## contains the error message. No exceptions @@ -151,13 +218,13 @@ type # implementation const - SymChars = {'a'..'z', 'A'..'Z', '0'..'9', '_', '\x80'..'\xFF', '.', '/', '\\', '-'} + SymChars = {'a'..'z', 'A'..'Z', '0'..'9', '_', ' ', '\x80'..'\xFF', '.', '/', '\\', '-'} proc rawGetTok(c: var CfgParser, tok: var Token) {.gcsafe.} proc open*(c: var CfgParser, input: Stream, filename: string, lineOffset = 0) {.rtl, extern: "npc$1".} = - ## initializes the parser with an input stream. `Filename` is only used + ## Initializes the parser with an input stream. `Filename` is only used ## for nice error messages. `lineOffset` can be used to influence the line ## number information in the generated error messages. lexbase.open(c, input) @@ -168,35 +235,21 @@ proc open*(c: var CfgParser, input: Stream, filename: string, rawGetTok(c, c.tok) proc close*(c: var CfgParser) {.rtl, extern: "npc$1".} = - ## closes the parser `c` and its associated input stream. + ## Closes the parser `c` and its associated input stream. lexbase.close(c) proc getColumn*(c: CfgParser): int {.rtl, extern: "npc$1".} = - ## get the current column the parser has arrived at. + ## Gets the current column the parser has arrived at. result = getColNumber(c, c.bufpos) proc getLine*(c: CfgParser): int {.rtl, extern: "npc$1".} = - ## get the current line the parser has arrived at. + ## Gets the current line the parser has arrived at. result = c.lineNumber proc getFilename*(c: CfgParser): string {.rtl, extern: "npc$1".} = - ## get the filename of the file that the parser processes. + ## Gets the filename of the file that the parser processes. result = c.filename -proc handleHexChar(c: var CfgParser, xi: var int) = - case c.buf[c.bufpos] - of '0'..'9': - xi = (xi shl 4) or (ord(c.buf[c.bufpos]) - ord('0')) - inc(c.bufpos) - of 'a'..'f': - xi = (xi shl 4) or (ord(c.buf[c.bufpos]) - ord('a') + 10) - inc(c.bufpos) - of 'A'..'F': - xi = (xi shl 4) or (ord(c.buf[c.bufpos]) - ord('A') + 10) - inc(c.bufpos) - else: - discard - proc handleDecChars(c: var CfgParser, xi: var int) = while c.buf[c.bufpos] in {'0'..'9'}: xi = (xi * 10) + (ord(c.buf[c.bufpos]) - ord('0')) @@ -241,8 +294,10 @@ proc getEscapedChar(c: var CfgParser, tok: var Token) = of 'x', 'X': inc(c.bufpos) var xi = 0 - handleHexChar(c, xi) - handleHexChar(c, xi) + if handleHexChar(c.buf[c.bufpos], xi): + inc(c.bufpos) + if handleHexChar(c.buf[c.bufpos], xi): + inc(c.bufpos) add(tok.literal, chr(xi)) of '0'..'9': var xi = 0 @@ -306,6 +361,10 @@ proc getSymbol(c: var CfgParser, tok: var Token) = add(tok.literal, c.buf[pos]) inc(pos) if not (c.buf[pos] in SymChars): break + + while tok.literal.len > 0 and tok.literal[^1] == ' ': + tok.literal.setLen(tok.literal.len - 1) + c.bufpos = pos tok.kind = tkSymbol @@ -367,19 +426,19 @@ proc rawGetTok(c: var CfgParser, tok: var Token) = else: getSymbol(c, tok) proc errorStr*(c: CfgParser, msg: string): string {.rtl, extern: "npc$1".} = - ## returns a properly formatted error message containing current line and + ## Returns a properly formatted error message containing current line and ## column information. result = `%`("$1($2, $3) Error: $4", [c.filename, $getLine(c), $getColumn(c), msg]) proc warningStr*(c: CfgParser, msg: string): string {.rtl, extern: "npc$1".} = - ## returns a properly formatted warning message containing current line and + ## Returns a properly formatted warning message containing current line and ## column information. result = `%`("$1($2, $3) Warning: $4", [c.filename, $getLine(c), $getColumn(c), msg]) proc ignoreMsg*(c: CfgParser, e: CfgEvent): string {.rtl, extern: "npc$1".} = - ## returns a properly formatted warning message containing that + ## Returns a properly formatted warning message containing that ## an entry is ignored. case e.kind of cfgSectionStart: result = c.warningStr("section ignored: " & e.section) @@ -393,7 +452,7 @@ proc getKeyValPair(c: var CfgParser, kind: CfgEventKind): CfgEvent = if c.tok.kind == tkSymbol: case kind of cfgOption, cfgKeyValuePair: - result = CfgEvent(kind: kind, key: c.tok.literal, value: "") + result = CfgEvent(kind: kind, key: c.tok.literal.move, value: "") else: discard rawGetTok(c, c.tok) if c.tok.kind in {tkEquals, tkColon}: @@ -410,7 +469,7 @@ proc getKeyValPair(c: var CfgParser, kind: CfgEventKind): CfgEvent = rawGetTok(c, c.tok) proc next*(c: var CfgParser): CfgEvent {.rtl, extern: "npc$1".} = - ## retrieves the first/next event. This controls the parser. + ## Retrieves the first/next event. This controls the parser. case c.tok.kind of tkEof: result = CfgEvent(kind: cfgEof) @@ -422,7 +481,7 @@ proc next*(c: var CfgParser): CfgEvent {.rtl, extern: "npc$1".} = of tkBracketLe: rawGetTok(c, c.tok) if c.tok.kind == tkSymbol: - result = CfgEvent(kind: cfgSectionStart, section: c.tok.literal) + result = CfgEvent(kind: cfgSectionStart, section: c.tok.literal.move) else: result = CfgEvent(kind: cfgError, msg: errorStr(c, "symbol expected, but found: " & c.tok.literal)) @@ -439,17 +498,17 @@ proc next*(c: var CfgParser): CfgEvent {.rtl, extern: "npc$1".} = # ---------------- Configuration file related operations ---------------- type - Config* = OrderedTableRef[string, <//>OrderedTableRef[string, string]] + Config* = OrderedTableRef[string, OrderedTableRef[string, string]] proc newConfig*(): Config = - ## Create a new configuration table. + ## Creates a new configuration table. ## Useful when wanting to create a configuration file. - result = newOrderedTable[string, <//>OrderedTableRef[string, string]]() + result = newOrderedTable[string, OrderedTableRef[string, string]]() -proc loadConfig*(stream: Stream, filename: string = "[stream]"): <//>Config = - ## Load the specified configuration from stream into a new Config instance. +proc loadConfig*(stream: Stream, filename: string = "[stream]"): Config = + ## Loads the specified configuration from stream into a new Config instance. ## `filename` parameter is only used for nicer error messages. - var dict = newOrderedTable[string, <//>OrderedTableRef[string, string]]() + var dict = newOrderedTable[string, OrderedTableRef[string, string]]() var curSection = "" ## Current section, ## the default value of the current section is "", ## which means that the current section is a common @@ -479,8 +538,8 @@ proc loadConfig*(stream: Stream, filename: string = "[stream]"): <//>Config = close(p) result = dict -proc loadConfig*(filename: string): <//>Config = - ## Load the specified configuration file into a new Config instance. +proc loadConfig*(filename: string): Config = + ## Loads the specified configuration file into a new Config instance. let file = open(filename, fmRead) let fileStream = newFileStream(file) defer: fileStream.close() @@ -505,9 +564,9 @@ proc replace(s: string): string = result = d proc writeConfig*(dict: Config, stream: Stream) = - ## Writes the contents of the table to the specified stream + ## Writes the contents of the table to the specified stream. ## - ## **Note:** Comment statement will be ignored. + ## .. note:: Comment statement will be ignored. for section, sectionData in dict.pairs(): if section != "": ## Not general section if not allCharsInSet(section, SymChars): ## Non system character @@ -546,7 +605,8 @@ proc writeConfig*(dict: Config, stream: Stream) = proc `$`*(dict: Config): string = ## Writes the contents of the table to string. - ## Note: Comment statement will be ignored. + ## + ## .. note:: Comment statement will be ignored. let stream = newStringStream() defer: stream.close() dict.writeConfig(stream) @@ -554,21 +614,23 @@ proc `$`*(dict: Config): string = proc writeConfig*(dict: Config, filename: string) = ## Writes the contents of the table to the specified configuration file. - ## Note: Comment statement will be ignored. + ## + ## .. note:: Comment statement will be ignored. let file = open(filename, fmWrite) defer: file.close() let fileStream = newFileStream(file) dict.writeConfig(fileStream) -proc getSectionValue*(dict: Config, section, key: string): string = - ## Gets the Key value of the specified Section, returns an empty string if the key does not exist. +proc getSectionValue*(dict: Config, section, key: string, defaultVal = ""): string = + ## Gets the key value of the specified Section. + ## Returns the specified default value if the specified key does not exist. if dict.hasKey(section): if dict[section].hasKey(key): result = dict[section][key] else: - result = "" + result = defaultVal else: - result = "" + result = defaultVal proc setSectionKey*(dict: var Config, section, key, value: string) = ## Sets the Key value of the specified Section. @@ -583,10 +645,15 @@ proc delSection*(dict: var Config, section: string) = dict.del(section) proc delSectionKey*(dict: var Config, section, key: string) = - ## Delete the key of the specified section. + ## Deletes the key of the specified section. if dict.hasKey(section): if dict[section].hasKey(key): if dict[section].len == 1: dict.del(section) else: dict[section].del(key) + +iterator sections*(dict: Config): lent string {.since: (1, 5).} = + ## Iterates through the sections in the `dict`. + for section in dict.keys: + yield section |