Add API tests for NamedDestsByName, DestGetPageIndex.

Add API tests: NamedDestsByName, DestGetPageIndex.

This also adds a fpdfdoc_embeddertest.cpp to keep the test
file name matching with the API call under test.

R=bo_xu@foxitsoftware.com

Committed: https://pdfium.googlesource.com/pdfium/+/83f96a01a2cfd64d0a370e74c3f4353eb9a102d2

[Tom Sepez, 2015-01-21 23:13:53]

tsepez@chromium.org changed reviewers:
+ bo_xu@foxitsoftware.com

[Tom Sepez, 2015-01-21 23:13:53]

Bo, please review.

[Bo Xu, 2015-01-22 00:28:52]

We are trying to get page 11, but there is only one page. It maybe good that we
have a document with those pages and page number rendered on each page and the
destinations can show up in the reader.

I can put it up a document like that.

https://codereview.chromium.org/812933004/diff/1/testing/resources/named_dest...
File testing/resources/named_dests.pdf (right):

https://codereview.chromium.org/812933004/diff/1/testing/resources/named_dest...
testing/resources/named_dests.pdf:60: /FirstAlternate [11 /XYZ 200 400 800]
11 here should be an indirect object according to pdf reference 8.2.1.

Let me check our developers to make sure the document conforms to the standard

[Tom Sepez, 2015-01-22 00:33:35]

On 2015/01/22 00:28:52, Bo Xu wrote:
> We are trying to get page 11, but there is only one page. It maybe good that
we
> have a document with those pages and page number rendered on each page and the
> destinations can show up in the reader.
> 
> I can put it up a document like that.
Actually, this is more interesting in some sense in that it forces us to ask
what the correct behaviour should be when the dest doesn't reference an actual
page.  Let me know how much more bulk this adds to a document.

> 
>
https://codereview.chromium.org/812933004/diff/1/testing/resources/named_dest...
> File testing/resources/named_dests.pdf (right):
> 
>
https://codereview.chromium.org/812933004/diff/1/testing/resources/named_dest...
> testing/resources/named_dests.pdf:60: /FirstAlternate [11 /XYZ 200 400 800]
> 11 here should be an indirect object according to pdf reference 8.2.1.
> 
> Let me check our developers to make sure the document conforms to the standard

[Tom Sepez, 2015-01-22 00:43:47]

> Let me check our developers to make sure the document conforms to the standard

Looks like you've got special code for this case, eg. doc_basic.cpp:19

18	    if (pPage->GetType() == PDFOBJ_NUMBER) {
19	        return pPage->GetInteger();
20	    }
21	    if (pPage->GetType() != PDFOBJ_DICTIONARY) {
22	        return 0;
23	    }
24	    return pDoc->GetPageIndex(pPage->GetObjNum());

Would be good to test both of these, esp with an invalid obj ref.

[Tom Sepez, 2015-01-22 01:08:52]

See PS#3.

[Bo Xu, 2015-01-22 17:59:32]

Wrong parents will cause adobe reader report error upon opening the pdf. (I know
I am always inclined to test them out in adobe readers:))

https://codereview.chromium.org/812933004/diff/60001/testing/resources/named_...
File testing/resources/named_dests.in (right):

https://codereview.chromium.org/812933004/diff/60001/testing/resources/named_...
testing/resources/named_dests.in:23: /Parent 1 0 R
should be "/Parent 2 0 R"

https://codereview.chromium.org/812933004/diff/60001/testing/resources/named_...
testing/resources/named_dests.in:33: /Parent 1 0 R
Ditto

[Tom Sepez, 2015-01-22 18:22:17]

On 2015/01/22 17:59:32, Bo Xu wrote:
> Wrong parents will cause adobe reader report error upon opening the pdf. (I
know
> I am always inclined to test them out in adobe readers:))
> 
>
https://codereview.chromium.org/812933004/diff/60001/testing/resources/named_...
> File testing/resources/named_dests.in (right):
> 
>
https://codereview.chromium.org/812933004/diff/60001/testing/resources/named_...
> testing/resources/named_dests.in:23: /Parent 1 0 R
> should be "/Parent 2 0 R"
> 
>
https://codereview.chromium.org/812933004/diff/60001/testing/resources/named_...
> testing/resources/named_dests.in:33: /Parent 1 0 R
> Ditto
Fixed.

[Bo Xu, 2015-01-22 19:06:17]

LGTM

[Tom Sepez, 2015-01-22 19:11:49]

Committed patchset #5 (id:80001) manually as
83f96a01a2cfd64d0a370e74c3f4353eb9a102d2 (presubmit successful).