Hi Folks,
I don't seem to be getting the correct results. Consider the following test on
the input hex 59 (0101 1001):
<xs:element name="Test_leastSignificantBitFirst">
<xs:complexType>
<xs:sequence
dfdl:bitOrder="leastSignificantBitFirst"
dfdl:lengthUnits="bits"
dfdl:alignment="1"
dfdl:alignmentUnits="bits"
dfdl:representation="binary"
dfdl:binaryNumberRep="binary">
<xs:element name="Three-bits" type="unsignedint3" />
<xs:element name="Two-bits" type="unsignedint2" />
<xs:element name="Three-bits" type="unsignedint3" />
</xs:sequence>
</xs:complexType>
</xs:element>
Running that results in this error message:
[error] Parse Error: Insufficient bits in data. Needed 2 bit(s).
Well, okay, so I commented out the Two-bits element. I re-ran it and got this
error message:
[error] Parse Error: Insufficient bits in data. Needed 3 bit(s).
Okay, so I comment out the last two elements, and re-ran it. No error this
time. But I did get a "Left over data" message. But the thing that really
concerns me is the output that was generated:
<Test_leastSignificantBitFirst>
<Three-bits>2</Three-bits>
</Test_leastSignificantBitFirst>
That's not correct. The value of <Three-bits> should be 1 (binary 001). Right?
What am I doing wrong and still not understanding?
/Roger
From: Beckerle, Mike <[email protected]>
Sent: Friday, February 15, 2019 11:45 AM
To: [email protected]
Subject: [EXT] Re: I don't understand bitOrder - please help!
This bitOrder feature is very thoroughly tested, so I suspect your tests are
missing something important.
dfdl:lengthUnits="bits"
dfdl:alignment="1"
dfdl:alignmentUnits="bits"
dfdl:representation="binary"
dfdl:binaryNumberRep="binary"
Those properties are really essential.
________________________________
From: Costello, Roger L. <[email protected]<mailto:[email protected]>>
Sent: Friday, February 15, 2019 11:14:18 AM
To: [email protected]<mailto:[email protected]>
Subject: Re: I don't understand bitOrder - please help!
Thanks Steve and David!
To follow up on Steve's example ...
The input is: 01011001
With this element declaration:
<xs:element name="bit3"
type="xs:int"
dfdl:length="3"
dfdl:bitOrder="mostSignificantBitFirst"... />
The value of <bit3> will be binary 010 (decimal 2), right?
With this element declaration:
<xs:element name="bit3"
type="xs:int"
dfdl:length="3"
dfdl:bitOrder="leastSignificantBitFirst"... />
The value of <bit3> will be binary 001 (decimal 1), right?
That is not the behavior that I got with my tests.
/Roger
-----Original Message-----
From: Steve Lawrence <[email protected]<mailto:[email protected]>>
Sent: Friday, February 15, 2019 8:50 AM
To: [email protected]<mailto:[email protected]>; David Winant
<[email protected]<mailto:[email protected]>>
Subject: Re: [EXT] Re: I don't understand bitOrder - please help!
That's correct. As a minimal example, let's imagine we had one byte of data
like so, with bits labels A-H:
Label: ABCDEFGH
Bits: 01011001
Imagine we had an element like so:
<xs:element name="bit3" type="xs:int" dfdl:length="3" ... />
So an integer with length 3 bits.
With bitOrder = MSBF, we would read the 3 most significant bits first.
So bits A-C and a value of 2 (010). If there was a following element, the next
bit read would be bit D.
With bitOrdre = LSBF, we would read the 3 least significant bits first.
So bits F-H and a value of 1 (001). If there was a following element, the next
bit read would be bit E.
On 2/15/19 8:33 AM, David Winant wrote:
> Roger,
>
> "Most significant" is implied as leftmost because that is the way we
> write numbers. It means the bit associated with the highest power of
> 2, 2^7 in the case of a byte.
>
> What the bitOrder controls is whether the "first" bit is read from the
> right or read from the left.
>
> David
>
> On Fri, Feb 15, 2019 at 8:13 AM Costello, Roger L. <[email protected]
<mailto:[email protected]%20%0b>> <mailto:[email protected]>> wrote:
>
> Hi Mike,____
>
> __ __
>
> * Least significant bit first or LSBF, the first bit is the rightmost
> zero.____
>
> __ __
>
> So the most significant bit is the leftmost one.____
>
> __ __
>
> * Most significant bit first, MSBF, the first bit is the 1 on
> the left.____
>
> __ __
>
> So the least significant bit is the rightmost zero.____
>
> __ __
>
> Aren't they saying the same thing: The most significant bit is the
> leftmost
> one and the least significant bit if the rightmost zero. What's the
> value/benefit/use-case of specifying bitOrder? ____
>
> __ __
>
> /Roger____
>
> __ __
>
> *From:* Beckerle, Mike <[email protected]
> <mailto:[email protected]>>
> *Sent:* Friday, February 15, 2019 8:02 AM
> *To:* [email protected]<mailto:[email protected]>
> <mailto:[email protected]>
> *Subject:* [EXT] Re: I don't understand bitOrder - please
> help!____
>
> __ __
>
> Probably bit order should have been called bit numbering system.
> ____
>
> It does nothing for whole bytes.____
>
> It is only about which bits are considered to occupy which index
> positions. ____
>
> So given byte with value 128, base 2 that is 80 hex or 10000000. So is the
> "first" bit a 1 or a zero?____
>
> Least significant bit first or LSBF, the first bit is the rightmost zero.
> Most significant bit first, MSBF, the first bit is the 1 on the
> left. ____
>
> Many format standards describe the bits numbering from right to left.
>
>
>
>
> ____
>
> Get Outlook for Android <https://aka.ms/ghei36>____
>
> __ __
>
>
> ----------------------------------------------------------------------
> ----------
>
> *From:*Costello, Roger L. <[email protected] <mailto:[email protected]>>
> *Sent:* Friday, February 15, 2019 7:33:29 AM
> *To:* [email protected]<mailto:[email protected]>
> <mailto:[email protected]>
> *Subject:* I don't understand bitOrder - please help! ____
>
> ____
>
> Hello DFDL community,____
>
> ____
>
> I've read the portions of the specification that discuss bitOrder.
> I'm lost.____
>
> ____
>
> I ran some tests to get a feel for the difference between
> leastSignificantBitFirst and mostSignificantBitFirst. See below. Based on
> my
> tests, it appears there is no difference between leastSignificantBitFirst
> and mostSignificantBitFirst. Would you explain what this bitOrder stuff is
> all about, please? /Roger____
>
> ____
>
> ____
>
>
>
> --
>
> David Winant |*Trident Systems Incorporated *Sr. Software Engineer,
> Integrated Systems Engineering Group
> 10201 Fairfax Boulevard
> <https://maps.google.com/?q=10201+Fairfax+Boulevard+%7C+Suite+300+%7C+
<https://maps.google.com/?q=10201+Fairfax+Boulevard+%7C+Suite+300+%7C+%0b>>
Fairfax,+VA+22030&entry=gmail&source=g> | Suite 300
> <https://maps.google.com/?q=10201+Fairfax+Boulevard+%7C+Suite+300+%7C+
<https://maps.google.com/?q=10201+Fairfax+Boulevard+%7C+Suite+300+%7C+%0b>>
Fairfax,+VA+22030&entry=gmail&source=g> | Fairfax, VA 22030
> <https://maps.google.com/?q=10201+Fairfax+Boulevard+%7C+Suite+300+%7C+
<https://maps.google.com/?q=10201+Fairfax+Boulevard+%7C+Suite+300+%7C+%0b>>
Fairfax,+VA+22030&entry=gmail&source=g>
>
> d: 703.267.6014 | c: 304.237.4960 | f: 703.359.6226
> e: [email protected]<mailto:[email protected]>
> <mailto:[email protected]> |
> www.tridsys.com<http://www.tridsys.com> <http://www.tridsys.com/>
>
> **
>
> 143x59-trident-logo-web-small
>
> /Notice: The information contained in this email message is considered
> confidential and proprietary to the sender and is intended solely for
> review and use by the named recipient. Any unauthorized review, use
> or distribution is strictly prohibited. If you have received this
> message in error, please advise the sender by reply email and delete
> the message./
>
