Re: Is.Serializable
"Simone Busoli" <[email protected]> Sun, 7 Sep 2008 09:23:44 +0200
| Newsgroups | gmane.comp.windows.dotnet.nunit.user |
|---|---|
| Message-ID | <[email protected]> |
--===============1333766821== Content-Type: multipart/alternative; boundary="----=_Part_64456_13727414.1220772224336" ------=_Part_64456_13727414.1220772224336 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline Hi Charlie, I'd prefer to work on it again, there's been some changes since I wrote it in the beginning, and it would be useful for me to get comfortable with them. If you agree, I think this would be the right moment to publish the list of steps required to create a new constraint. On Sun, Sep 7, 2008 at 1:33 AM, Charlie Poole <[email protected]>wrote: > Hi Simone, > > I'm now back to looking at this patch, which I set aside while working on > other stuff. > > I agree with you about having two different constraints. Do you want me to > put in > the patch you sent and change the name to "Binary..." or would you like to > rework it? > > As noted, we do now have a generic AttributeConstraint. > > Charlie > > ------------------------------ > *From:* Simone Busoli [mailto:[email protected]] > *Sent:* Thursday, August 14, 2008 11:15 AM > *To:* Charlie Poole > *Cc:* [email protected] > > *Subject:* Re: [Nunit-users] Is.Serializable > > >> To check just for the attribute or the interface >> Is.Serializable() OR >> Is.Serializable >> >> To check by trying to do it >> Is.Serializable(formatter) OR >> Is.Serializable.Using(formatter) >> > > I think that making Is.Serializable check for the presence of the attribute > on the class might be misleading, because its presence doesn't mean it's > serializable, and its absence doesn't mean it isn't. I think that > Is.Serializable should be used to check that an object is really > serializable, together with its references. That is, Is.Serializable should > come in two flavours, one that checks if it is serializable in binary > format, and another for xml. I'll come to this later. > About checking for the presence of the attribute, as you mentioned in a > comment to Fabio's feature request, I think it could be generalized with > Has.Attribute<> or Has.Attribute(Type). > So, given that the presence of an attribute is checked with a generic > Has.Attribute, the way I would tackle the true ability of an object graph to > serialize is either using Is.XmlSerializable and Is.BinarySerializable or > something equivalent. I don't get how you could pass in an instance of a > formatter; I guess you proposed it because you might want to be able to use > the assertion for other types of formatters not built into the .net fx s > well, but then how would you call their methods? Via reflection? > > >> >> The aternatives above are between using a method and using a property. It >> would be >> cool to use the property for te first example and the method for the >> second, but >> that's not possible. >> >> Most of our syntactic elements use a property format when there is no >> argument. >> When an element has an optional argument, then we are forced to use method >> syntax, which may look a bit odd. For example, we have to useIs.Ordered() >> rather than Is.Ordered - That example should be familiar to you :-) >> >> I lean toward "Using" or "With" or some other keyword. >> >> What do folks think? >> > > I'm fine with anything you choose. I don't think it's so important if it's > a property or a method, just let's try to be consistent :) > > >> >> Charlie >> >> ------------------------------ >> *From:* Simone Busoli [mailto:[email protected]] >> *Sent:* Thursday, August 14, 2008 12:16 AM >> *To:* Charlie Poole >> *Subject:* Re: [Nunit-users] Is.Serializable >> >> Hmm, I just realized that what I'm doing is not enough for the object >> to be serializable. I noticed that the prerequisites for it to be >> serializable are quite different according to whether you want to serialize >> it to xml or to binary. For example, the binary serialization doesn't >> require classes to be public, while xml does. xml serialization doesn't >> require the serializable attribute, while the binary does. So I'm wondering >> if it could be the case to split the assertion in IsBinarySerializable and >> IsXmlSerializable, or check that it is both binary and xml serializable in >> the same assertion. Thoughts? >> >> On Thu, Aug 14, 2008 at 2:02 AM, Charlie Poole < >> [email protected]> wrote: >> >>> Hi Simone, >>> >>> This looks good to me. Would you like to synch up and fix it so it works >>> with the latest code? My major >>> changes to Constraint, ConstraintBuilder, etc. are done and it would give >>> you a good way for you >>> to get familiar with those changes. Once it's done, you can either send >>> me a new patch or - if >>> you would like - I'll make you a commiter so you can add it yourself. If >>> you agree, I'll explain more >>> to you about review of code etc. >>> >>> Charlie >>> >>> ------------------------------ >>> *From:* [email protected] [mailto: >>> [email protected]] *On Behalf Of *Simone Busoli >>> *Sent:* Monday, August 11, 2008 7:11 AM >>> *To:* [email protected] >>> *Subject:* [Nunit-users] Is.Serializable >>> >>> With regards to: >>> http://sourceforge.net/tracker/index.php?func=detail&aid=2028517&group_id=10749&atid=360749 >>> >>> I have implemented a SerializableConstraint accessible via >>> Is.Serializable. The match is true if serialization to and deserialization >>> from a memory stream via a BinaryFormatter both succeed and the deserialized >>> object is not null. Otherwise the match returns false. >>> >>> >> > ------=_Part_64456_13727414.1220772224336 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline <div dir="ltr">Hi Charlie,<div><br></div><div>I'd prefer to work on it again, there's been some changes since I wrote it in the beginning, and it would be useful for me to get comfortable with them. If you agree, I think this would be the right moment to publish the list of steps required to create a new constraint.<br> <br><div class="gmail_quote">On Sun, Sep 7, 2008 at 1:33 AM, Charlie Poole <span dir="ltr"><<a href="mailto:[email protected]">[email protected]</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"> <div> <div><span><font face="Arial" color="#0000ff" size="2">Hi Simone,</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">I'm now back to looking at this patch, which I set aside while working on other stuff.</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">I agree with you about having two different constraints. Do you want me to put in</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">the patch you sent and change the name to "Binary..." or would you like to rework it?</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">As noted, we do now have a generic AttributeConstraint.</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">Charlie</font></span></div><br> <blockquote dir="ltr" style="padding-left:5px;margin-left:5px;border-left:#0000ff 2px solid;margin-right:0px"> <div lang="en-us" dir="ltr" align="left"> <hr> <font face="Tahoma" size="2"><div class="Ih2E3d"><b>From:</b> Simone Busoli [mailto:<a href="mailto:[email protected]" target="_blank">[email protected]</a>] <br></div><b>Sent:</b> Thursday, August 14, 2008 11:15 AM<br><b>To:</b> Charlie Poole<br><b>Cc:</b> <a href="mailto:[email protected]" target="_blank">[email protected]</a><div><div></div><div class="Wj3C7c"><br><b>Subject:</b> Re: [Nunit-users] Is.Serializable<br></div></div></font><br></div><div><div></div><div class="Wj3C7c"> <div></div> <div dir="ltr"> <div class="gmail_quote"> <blockquote class="gmail_quote" style="padding-left:1ex;margin:0pt 0pt 0pt 0.8ex;border-left:rgb(204,204,204) 1px solid"> <div><br> <div><span><font face="Arial" color="#0000ff" size="2">To check just for the attribute or the interface</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"> Is.Serializable() OR</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"> Is.Serializable</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">To check by trying to do it</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"> Is.Serializable(formatter) OR</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"> Is.Serializable.Using(formatter)</font></span></div></div></blockquote> <div><br>I think that making Is.Serializable check for the presence of the attribute on the class might be misleading, because its presence doesn't mean it's serializable, and its absence doesn't mean it isn't. I think that Is.Serializable should be used to check that an object is really serializable, together with its references. That is, Is.Serializable should come in two flavours, one that checks if it is serializable in binary format, and another for xml. I'll come to this later.<br>About checking for the presence of the attribute, as you mentioned in a comment to Fabio's feature request, I think it could be generalized with Has.Attribute<> or Has.Attribute(Type).<br>So, given that the presence of an attribute is checked with a generic Has.Attribute, the way I would tackle the true ability of an object graph to serialize is either using Is.XmlSerializable and Is.BinarySerializable or something equivalent. I don't get how you could pass in an instance of a formatter; I guess you proposed it because you might want to be able to use the assertion for other types of formatters not built into the .net fx s well, but then how would you call their methods? Via reflection?<br> </div> <blockquote class="gmail_quote" style="padding-left:1ex;margin:0pt 0pt 0pt 0.8ex;border-left:rgb(204,204,204) 1px solid"> <div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">The aternatives above are between using a method and using a property. It would be</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">cool to use the property for te first example and the method for the second, but </font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">that's not possible.</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">Most of our syntactic elements use a property format when there is no argument.</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">When an element has an optional argument, then we are forced to use method</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">syntax, which may look a bit odd. For example, we have to use</font></span><span><font face="Arial" color="#0000ff" size="2"> Is.Ordered()</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">rather than Is.Ordered - That example should be familiar to you :-)</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">I lean toward "Using" or "With" or some other keyword.</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">What do folks think?</font></span></div></div></blockquote> <div><br>I'm fine with anything you choose. I don't think it's so important if it's a property or a method, just let's try to be consistent :)<br> </div> <blockquote class="gmail_quote" style="padding-left:1ex;margin:0pt 0pt 0pt 0.8ex;border-left:rgb(204,204,204) 1px solid"> <div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div><font color="#888888"> <div><span><font face="Arial" color="#0000ff" size="2">Charlie</font></span></div><br></font> <blockquote dir="ltr" style="padding-left:5px;margin-left:5px;border-left:rgb(0,0,255) 2px solid;margin-right:0px"> <div> <div lang="en-us" dir="ltr" align="left"> <hr> <font face="Tahoma" size="2"><b>From:</b> Simone Busoli [mailto:<a href="mailto:[email protected]" target="_blank">[email protected]</a>] <br><b>Sent:</b> Thursday, August 14, 2008 12:16 AM<br><b>To:</b> Charlie Poole<br><b>Subject:</b> Re: [Nunit-users] Is.Serializable<br></font><br></div> <div></div></div> <div> <div></div> <div> <div dir="ltr">Hmm, I just realized that what I'm doing is not enough for the object to be serializable. I noticed that the prerequisites for it to be serializable are quite different according to whether you want to serialize it to xml or to binary. For example, the binary serialization doesn't require classes to be public, while xml does. xml serialization doesn't require the serializable attribute, while the binary does. So I'm wondering if it could be the case to split the assertion in IsBinarySerializable and IsXmlSerializable, or check that it is both binary and xml serializable in the same assertion. Thoughts?<br><br> <div class="gmail_quote">On Thu, Aug 14, 2008 at 2:02 AM, Charlie Poole <span dir="ltr"><<a href="mailto:[email protected]" target="_blank">[email protected]</a>></span> wrote:<br> <blockquote class="gmail_quote" style="padding-left:1ex;margin:0pt 0pt 0pt 0.8ex;border-left:rgb(204,204,204) 1px solid"> <div> <div><span><font face="Arial" color="#0000ff" size="2">Hi Simone,</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2"></font></span> </div> <div><span><font face="Arial" color="#0000ff" size="2">This looks good to me. Would you like to synch up and fix it so it works with the latest code? My major</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">changes to Constraint, ConstraintBuilder, etc. are done and it would give you a good way for you</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">to get familiar with those changes. Once it's done, you can either send me a new patch or - if</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">you would like - I'll make you a commiter so you can add it yourself. If you agree, I'll explain more</font></span></div> <div><span><font face="Arial" color="#0000ff" size="2">to you about review of code etc.</font></span></div> <div> </div> <div><span></span><font face="Arial"><font color="#0000ff"><font size="2">C<span>harlie</span></font></font></font><br></div> <blockquote dir="ltr" style="padding-left:5px;margin-left:5px;border-left:rgb(0,0,255) 2px solid;margin-right:0px"> <div lang="en-us" dir="ltr" align="left"> <hr> <font face="Tahoma" size="2"><b>From:</b> <a href="mailto:[email protected]" target="_blank">[email protected]</a> [mailto:<a href="mailto:[email protected]" target="_blank">[email protected]</a>] <b>On Behalf Of </b>Simone Busoli<br><b>Sent:</b> Monday, August 11, 2008 7:11 AM<br><b>To:</b> <a href="mailto:[email protected]" target="_blank">[email protected]</a><br><b>Subject:</b> [Nunit-users] Is.Serializable<br></font><br></div> <div> <div></div> <div dir="ltr">With regards to: <a href="http://sourceforge.net/tracker/index.php?func=detail&aid=2028517&group_id=10749&atid=360749" target="_blank">http://sourceforge.net/tracker/index.php?func=detail&aid=2028517&group_id=10749&atid=360749</a><br> <br>I have implemented a SerializableConstraint accessible via Is.Serializable. The match is true if serialization to and deserialization from a memory stream via a BinaryFormatter both succeed and the deserialized object is not null. Otherwise the match returns false.<br></div></div></blockquote></div></blockquote></div><br></div></div></div></blockquote></div></blockquote></div><br></div></div></div></blockquote></div> </blockquote></div><br></div></div> ------=_Part_64456_13727414.1220772224336-- --===============1333766821== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ --===============1333766821== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Nunit-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/nunit-users --===============1333766821==--