…r rather than java.lang.Object
…treams protocol - added appropriate protocol implementations for byte and character arrays - eliminated PrintWriter returns from writer impls, write-lines, and spit - added copy implementations reading from character arrays Signed-off-by: Stuart Halloway <email@example.com>
…s in ~f and ~$. See #47.
* For backward compatibility with 1.1 release. * Namespaces copied: duck-streams, java-utils, seq-utils, shell-out, str-utils, str-utils2. * These namespaces can be marked as deprecated. * They are copied rather than simply aliased because some behavior has changed.
clojure.contrib.io already supports character-oriented I/O through the multi-methods reader and writer. This patch adds support for byte-oriented I/O by providing the multi-methods input-stream and output-stream. * input-stream knows how to open InputStreams for reading bytes. * reader has been refactored to build on input-stream. * output-stream knows how to open OutputStreams for writing bytes. * writer has been refactored to build on output-stream (where sensible) By recognizing that output-stream will throw exceptions for us if it's unable to open the underlying resource, we were able to use the :default method to cover URL, URI and Socket. The String writer has not been touched. (Writing it in terms of output-stream would have made it longer and more complex.) * *append-to-writer* has been renamed to *append* for use with output-streams without souding foolish. This is a breaking change for clients that bind *append-to-writer* instead of calling append-writer or append-split as recommended in the docsting. * Like writer, output-stream consults the current binding of *append*. * append-output-stream is analagous to append-writer. Like append-writer, the implementation looks very general, but in reality it only works for files. Signed-off-by: Stuart Halloway <firstname.lastname@example.org>
- fixed bug: misspelling in pom - updated gen-html-docs to track c.c.string name changes