On 17 Apr 2020, at 15:11, Roger Riggs <roger.ri...@oracle.com> wrote:
Hi Pavel,
On 4/17/20 8:27 AM, Pavel Rappo wrote:
Hi Roger,
First of all, many thanks for doing this. After skimming through that webrev I
have two comments.
1. It might make sense to beef this up with more retrofitted tests, so people
could more clearly see how applicable this API is. If you'd like I could help
you with looking for those across the OpenJDK codebase.
There are fewer (tests) than you might think though I only grepped for
HexDumpEncoder.
I omitted a few that are dependent on parsing HexDumpEncoder formatted strings
and there are golden files to be considered.
A separate pass will be needed for those.
On 17 Apr 2020, at 15:11, Roger Riggs <roger.ri...@oracle.com> wrote:
Hi Pavel,
On 4/17/20 8:27 AM, Pavel Rappo wrote:
Hi Roger,
First of all, many thanks for doing this. After skimming through that webrev I
have two comments.
1. It might make sense to beef this up with more retrofitted tests, so people
could more clearly see how applicable this API is. If you'd like I could help
you with looking for those across the OpenJDK codebase.
There are fewer (tests) than you might think though I only grepped for
HexDumpEncoder.
I omitted a few that are dependent on parsing HexDumpEncoder formatted strings
and there are golden files to be considered.
A separate pass will be needed for those.
I didn't consider any uses in the src/... tree since at the moment this is only
considered for testing. If it passes muster, it deserves further consideration
for JDK.
2. Correct me if I'm wrong, but I couldn't find the complementary functionality
of parsing in a hexdump in that webrev. I'm not sure how prevalent this is, but
I can attest that there is an occasional need for it. For instance, when
implementing the HPACK format I coded the tests from that RFC which use a
two-byte hexadecimal representation. For example,
1008 7061 7373 776f 7264 0673 6563 7265 | ..password.secre
74 | t
I'm sure that's not the only case where hex parsing might be needed.
I developed a corresponding parser but it needs more work.
Thanks, Roger
Regardless of whether or not this functionality is already included in that
webrev (I might've missed it), the hex printing in and of itself is valuable.
-Pavel
On 16 Apr 2020, at 20:17, Roger Riggs <roger.ri...@oracle.com> wrote:
Please review[2] and comment on a new Hex printing utility to support OpenJDK
tests.
The usefulness of a hex printing has been discussed from time to time with
many suggestions as to the API shape and features.
To get an idea of the API and features, take a look at the javadoc[1].
It covers the basic formatting of offset, and values, and extends the
descriptive part beyond simple ASCII or printable so that a custom formatter
can interpret the byte stream in parallel to the bytes.
The webrev includes changes to existing tests that currently
use HexDumpEncoder to use HexPrinter.
Comments appreciated, Roger
[1]Javadoc:
http://cr.openjdk.java.net/~rriggs/webrev-hexprinter-8243010/javadoc
[2] Webrev:
http://cr.openjdk.java.net/~rriggs/webrev-hexprinter-8243010/
[3] Issue:
https://bugs.openjdk.java.net/browse/JDK-8243010
I didn't consider any uses in the src/... tree since at the moment this is only
considered for testing. If it passes muster, it deserves further consideration
for JDK.
2. Correct me if I'm wrong, but I couldn't find the complementary functionality
of parsing in a hexdump in that webrev. I'm not sure how prevalent this is, but
I can attest that there is an occasional need for it. For instance, when
implementing the HPACK format I coded the tests from that RFC which use a
two-byte hexadecimal representation. For example,
1008 7061 7373 776f 7264 0673 6563 7265 | ..password.secre
74 | t
I'm sure that's not the only case where hex parsing might be needed.
I developed a corresponding parser but it needs more work.
Thanks, Roger
Regardless of whether or not this functionality is already included in that
webrev (I might've missed it), the hex printing in and of itself is valuable.
-Pavel
On 16 Apr 2020, at 20:17, Roger Riggs <roger.ri...@oracle.com> wrote:
Please review[2] and comment on a new Hex printing utility to support OpenJDK
tests.
The usefulness of a hex printing has been discussed from time to time with
many suggestions as to the API shape and features.
To get an idea of the API and features, take a look at the javadoc[1].
It covers the basic formatting of offset, and values, and extends the
descriptive part beyond simple ASCII or printable so that a custom formatter
can interpret the byte stream in parallel to the bytes.
The webrev includes changes to existing tests that currently
use HexDumpEncoder to use HexPrinter.
Comments appreciated, Roger
[1]Javadoc:
http://cr.openjdk.java.net/~rriggs/webrev-hexprinter-8243010/javadoc
[2] Webrev:
http://cr.openjdk.java.net/~rriggs/webrev-hexprinter-8243010/
[3] Issue:
https://bugs.openjdk.java.net/browse/JDK-8243010
