move read_all, try_cat, and poll_in to PublicInbox::IO

The IO package seems like a better home for I/O subs than the
Git package.  We lose the 60 second read timeout for `git
cat-file --batch-*' processes since it's probably not necessary
given how reliable the code has proven and things would fall
over hard in other ways if the storage device were completely
hosed.

1 files changed, 22 insertions, 4 deletions

diff --git a/lib/PublicInbox/IO.pm b/lib/PublicInbox/IO.pm
index 4c92566d..0d303500 100644
--- a/lib/PublicInbox/IO.pm
+++ b/lib/PublicInbox/IO.pm
@@ -6,10 +6,9 @@ package PublicInbox::IO;
 use v5.12;
 use parent qw(IO::Handle Exporter);
 use PublicInbox::DS qw(awaitpid);
-our @EXPORT_OK = qw(write_file);
-
-# TODO: this can probably be the new home for read_all, try_cat
-# and maybe even buffered read/readline...
+our @EXPORT_OK = qw(poll_in read_all try_cat write_file);
+use Carp qw(croak);
+use IO::Poll qw(POLLIN);
 
 sub waitcb { # awaitpid callback
         my ($pid, $errref, $cb, @args) = @_;
@@ -59,4 +58,23 @@ sub write_file ($$@) { # mode, filename, LIST (for print)
         defined(wantarray) && !wantarray ? $fh : close $fh;
 }
 
+sub poll_in ($;$) {
+        IO::Poll::_poll($_[1] // -1, fileno($_[0]), my $ev = POLLIN);
+}
+
+sub read_all ($;$$) {
+        use autodie qw(read);
+        my ($io, $len, $bref) = @_;
+        $bref //= \(my $buf);
+        my $r = read($io, $$bref, $len //= -s $io);
+        croak("read($io) ($r != $len)") if $len != $r;
+        $$bref;
+}
+
+sub try_cat ($) {
+        my ($path) = @_;
+        open(my $fh, '<', $path) or return '';
+        read_all $fh;
+}
+
 1;