bend-kit-multipart

Releases (6)

• 0.3.0.0 indexed
• 0.2.0.1 unindexed
• 0.2.0.0 invalid

  expected a fresh constructor name (duplicate declaration: Text)

  at multipart.bend:324

  multipart.bend does not load

• 0.1.1.1 invalid

  expected a fresh constructor name (duplicate declaration: Text)

  at multipart.bend:324

  multipart.bend does not load

• 0.1.1.0 invalid

  expected a fresh constructor name (duplicate declaration: Text)

  at multipart.bend:324

  multipart.bend does not load

• 0.1.0.0 invalid

  expected a fresh constructor name (duplicate declaration: Text)

  at multipart.bend:324

  multipart.bend does not load

Definitions (99)

multipart.bend

• type PartOne form field.
• type ItemWhat the streaming decoder reports, in order: Head, then zero or more Body pieces, then Tail, per part.
• def split1.consThe text before the first c, and the text after it.
• def split1
• type ClsHeader parameters (RFC 2045 §5.1, RFC 7578 §4.2): key=token or key="quoted", split on ';'.
• type Pm
• type Param
• type PSmode, the key and value so far (reversed), and the parameters found (latest first).
• def cls
• def params.emit
• def params.step
• def params.end
• def params.go
• def paramsThe parameters of a header value's text after its first ';', in order.
• def lookupThe first parameter named k (lowercase).
• def esc.oneWHATWG form-data escapes for names and filenames.
• def esc
• def unesc.fixacc is reversed, so "%22" shows as "22%".
• def unesc.go
• def unescUndoes esc, as the Fetch spec's form-data parser does.
• def unesc.m
• def bcharRFC 2046 §5.1.1 bchars: DIGIT ALPHA ' ( ) + _ , - .
• def bchars
• def boundary.ok1 to 70 bchars, not ending in a space.
• def boundary.words
• def boundary.new"bend-kit-" and 128 bits from the OS secure random source, as hex: 41 bchars.
• def content_typeThe Content-Type header value for a body encoded with boundary b.
• def boundary.get.of
• def boundary.getThe boundary parameter of a Content-Type value, such as a request's, or None.
• def enc.fname
• def enc.ctype
• def enc.head
• def enc.ctype.okA Content-Type with CR or LF would inject header lines.
• def enc.found
• def enc.body
• def enc.partThe pieces so far, latest first.
• def enc.fin
• def enc.go
• def encode.if
• def encodeThe body for parts under boundary b (RFC 7578 §4.1).
• def form.enc
• def form.of
• def formThe Content-Type value and body for parts, under a fresh random boundary.
• type StagePre: before the first delimiter.
• type Step
• type Decdelim is CRLF "--" boundary and dlen its length; buf holds the bytes not yet decided.
• def MAX_HEADponytail: a fixed cap on a header block or boundary line; make it a decoder field if a caller needs another.
• def drop
• def keep.two
• def keep.of
• def keepb as (all but its last k bytes, its last k bytes).
• def emit
• def wait.if
• def wait.bigWait for more input, unless the undecided bytes already pass MAX_HEAD.
• def suffix.dashWhat follows a delimiter match (RFC 2046 §5.1.1): "--", or transport padding then CRLF, makes it a delimiter (0).
• def suffix.lf
• def suffix
• def suffix.winponytail: looks at 64 bytes past the match; padding longer than that counts as data.
• def suffix.of
• def suffix.atThe buffer and the kind of the match at i.
• def pre.ifPre: the preamble is dropped.
• def pre.kind
• def pre.hit
• def after.lineAfter: suffix already saw "--" or padding then CRLF, so both are in buf.
• def after.dash
• def meta.lineHeader block to a Head (RFC 7578 §4.2, §4.4).
• def meta.go
• def head.named
• def head.form
• def head.disp
• def head.meta
• def head.parseLines split on LF; trimming drops each CR.
• def heads.item
• def heads.cut
• def heads.foundbuf starts with the CRLF that ended the boundary line, so a match at 0 is an empty block.
• def text.flush.lenContent: bytes that cannot begin a delimiter go out as Body; the last dlen - 1 wait for the next chunk.
• def text.flush
• def text.cut
• def text.dataA match that is data goes out through its first byte, and the search goes on after it.
• def text.holdA match whose suffix has not arrived: the bytes before it go out, and it waits.
• def text.if
• def text.kind
• def text.hit
• def step
• def go.stop
• def goEach More step consumes a byte, except the two after a delimiter, which itself consumes dlen >= 5, so len + 8 steps suffice.
• def decoderA decoder for boundary b.
• def feed.go
• def feedThe next chunk of a body, in any split.
• def finishDone once the close delimiter has been read; the epilogue is ignored.
• def collect.add
• def collectItems folded into parts, latest first.
• def chunks.fed
• def chunks.step
• def chunks.fin
• def chunks.end
• def chunks.go
• def decode.chunksThe parts of a body that arrives as chunks, split anywhere.
• def decodeThe parts of a whole body.