Menu
▾
▴
Re: [Quantlib-dev] ORE 7
|
From: Ioannis R. <qua...@de...> - 2022-10-08 17:12:22
|
Hi Peter, I will explain my issue. My Deriscope software consists of custom code that makes use of methods implemented either in ORE or QuantLib. Until ORE release 6 I was able to upgrade to the latest QuantLib release without any problems. But now with the current ORE release 7 I am stuck with the forked QuantLib code in the ORE repository. For example, if tomorrow Luigi makes a new QuantLib release, I will not be able to incorporate it into my code as it would be incompatible with what ORE expects. So, right now everything seems to work, but I am mostly concerned about the future. Optimally, the next ORE release 8 will link to a main branch (official) QuantLib version rather than a forked one. I can only hope this will be technically possible. Regards, Ioannis On 10/8/2022 5:07 PM, Peter Caspers wrote: > Hi Ioannis > > there is definitely no such thing as a "divorce from QuantLib" or any > other change in our dev strategy with release 7. That would be rather > silly indeed. Quite on the contrary I mentioned in a recent post that > it would make a lot of sense to consolidate some QuantExt classes with > QuantLib to reduce redundancies in (for example) the short rate models > (in the discussion around the Hull White 1F / GSR / LGM models, which > are all essentially the same). > > What is the concrete issue you have? Can you elaborate a bit, maybe we > can help mitigating it until the libraries are more in line again. I > am actually aware of a breaking change in the inflation framework, > this was reported on github recently. This is something we should > surely try to straighten out relatively soon. > > Thanks > Peter > > > On Sat, 8 Oct 2022 at 10:42, Ioannis Rigopoulos > <qua...@de...> wrote: > > Hi Roland, Peter, > > I am personally very keen on the continuous success of both > QuantLib and ORE as I am using both in my Deriscope application, > which is currently deployed successfully to a few bank clients, > including trading desks. > > I have opted to link to both QuantLib and ORE because of their > distinct competencies. > > Through the interaction with my clients, I often feel the need to > edit certain code segments. In the past I tried to add my edits to > the QuantLib and ORE main branches, but I have now refrained from > this practice due to a) QuantLib team's "less-than-enthusiastic" > response on issues I was raising and b) ORE's recent "disappearance". > > I felt very glad with ORE's comeback with release 7, its new > Acadia venture and its commitment to regular quarterly releases to > the point that I thought of transferring all my own code edits to > the ORE main branch. > > But I see now that the ORE release 7 marks a kind of "divorce" > from QuantLib, which is not good news for developers like me. > > To put it simply, I am now forced to choose one or the other: > > If I choose ORE, I will be distancing myself from the original > QuantLib branch, which would be dangerous if tomorrow ORE stops > being developed and also prohibit me from using recent additions > to QuantLib that are not yet incorporated in the ORE releases. > > If I choose QuantLib, I will not be able to use ORE, which > admittedly has very important stuff! > > A bit of an unfortunate deadlock really, which was not there until > release 6. > > Ioannis > > On 10/8/2022 9:49 AM, Peter Caspers wrote: >> Hi Ioannis >> >> we try to keep the diff between the original QL and our fork as >> small as possible. These particular differences refer to >> >> https://github.com/lballabio/QuantLib/pull/1222 >> >> which was needed to resolve major performance problems on our >> side. The discussion went a bit stale admittedly. I'll try to >> reignite it and flesh out the suggestions in the thread a bit more. >> >> Thank you >> Peter >> >> >> >> On Fri, 7 Oct 2022 at 19:29, Ioannis Rigopoulos >> <qua...@de...> wrote: >> >> Hi Roland, >> >> I have managed to compile and build ORE using the forked >> QuantLib version as suggested, but now I am stumbling on >> deep-rooted run time errors. >> >> It seems to me that the forked version of QuantLib deviates >> from the original in quite deep and subtle ways to the point >> that the rest of my code that used to work well with both ORE >> and QuantLib, now behaves in unexpected ways. >> >> One example is the very important QuantLib class >> FloatingRateCoupon , of which the original declaration >> >> class FloatingRateCoupon : public Coupon, public Observer >> >> has been changed to >> >> class FloatingRateCoupon : public Coupon, public LazyObject >> >> Another example that affected my run time behavior of OIS >> curve building concerns the implementation of >> OISRateHelper::impliedQuote(). >> >> The original QuantLib implementation contains the statement >> >> swap_->recalculate(); >> >> The forked version has replaced the above with >> >> swap_->deepUpdate(); >> >> It thus seems that the ORE version 7 marks a significant >> break from its previous versions as it requires a profoundly >> different underlying QuantLib code. >> >> I am sure that the ORE developers had good reasons for doing >> so, but the sad fact is that developers like me have to >> decide between using the original lballabio QuantLib branch >> or the one forked and maintained by ORE. >> >> Until ORE version 6, such a dilemma did not exist! >> >> Best regards, >> >> Ioannis >> >> On 10/4/2022 12:59 PM, Ioannis Rigopoulos wrote: >>> >>> Thanks for the quick turnaround. I have already extracted >>> the code using git, but it is nice to have this download >>> option as well. >>> >>> On 10/4/2022 12:46 PM, Roland Lichters wrote: >>>> >>>> Hi Ioannis, >>>> >>>> That’s a good point, we have added a release zip/tgz to the >>>> QuantLib repo in ORE: >>>> >>>> https://github.com/OpenSourceRisk/QuantLib/releases/tag/ORE-QuantLib-v1.27.1 >>>> >>>> I hope that works! >>>> >>>> Best regards, >>>> >>>> Roland >>>> >>>> *From: *Ioannis Rigopoulos <qua...@de...> >>>> <mailto:qua...@de...> >>>> *Date: *Tuesday, 4 October 2022 at 09:40 >>>> *To: *Roland Lichters <rol...@ac...c> >>>> <mailto:rol...@ac...c>, QuantLib users >>>> <qua...@li...> >>>> <mailto:qua...@li...>, QuantLib >>>> Mailing Lists <qua...@li...> >>>> <mailto:qua...@li...> >>>> *Subject: *Re: [Quantlib-dev] ORE 7 >>>> >>>> CAUTION:External email warning. This email originated from >>>> outside acadia.inc. Beware of phishing attempts. Comply >>>> with all acadia.inc policy to confirm sender identity >>>> before responding, forwarding, and especially clicking >>>> links or opening attachments. >>>> >>>> Thanks for the quick reply Roland. >>>> >>>> I will certainly try your suggestion, although I am using >>>> GitHub Desktop and I am not very familiar with the git dos >>>> commands. >>>> >>>> But since several people rely on the compressed forms (zip, >>>> tar) of the ORE and QuantLib releases, I am wondering if it >>>> would be possible to also provide the required QuantLib >>>> source code in compressed form, which is anyway done for >>>> all the official ORE and QuantLib releases. >>>> >>>> For example, you will see at the page >>>> https://github.com/OpenSourceRisk/Engine/releases/tag/v1.8.7.0 >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FOpenSourceRisk%2FEngine%2Freleases%2Ftag%2Fv1.8.7.0&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660004960075%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=3gnaMN97LTtKtZrRDG4GkT8op1tAABa%2BHjSWjvUZnz8%3D&reserved=0> >>>> the compressed ORE code. It would be certainly helpful if >>>> the corresponding QuantLib code could be similarly found >>>> somewhere, either as stand alone or embedded in the ORE >>>> compressed file. >>>> >>>> Ioannis >>>> >>>> On 10/4/2022 9:03 AM, Roland Lichters wrote: >>>> >>>> Hi Ioannis, >>>> >>>> Ah, maybe that wasn’t clear in the mail, we use a fork >>>> of QuantLib 1.27.1 with a few changes. >>>> >>>> So when you build ORE from sources on gihub, then >>>> please do a >>>> >>>> git submodule update --init >>>> >>>> That checks out our QuantLib 1.27.1 fork, and it should >>>> remove your build errors. >>>> >>>> Best regards, >>>> >>>> Roland >>>> >>>> *From: *Ioannis Rigopoulos <qua...@de...> >>>> <mailto:qua...@de...> >>>> *Date: *Monday, 3 October 2022 at 17:47 >>>> *To: *Roland Lichters <rol...@ac...c> >>>> <mailto:rol...@ac...c>, QuantLib users >>>> <qua...@li...> >>>> <mailto:qua...@li...>, QuantLib >>>> Mailing Lists <qua...@li...> >>>> <mailto:qua...@li...> >>>> *Subject: *Re: [Quantlib-dev] ORE 7 >>>> >>>> CAUTION:External email warning. This email originated >>>> from outside acadia.inc. Beware of phishing attempts. >>>> Comply with all acadia.inc policy to confirm sender >>>> identity before responding, forwarding, and especially >>>> clicking links or opening attachments. >>>> >>>> Thank you Roland for this new release, but I have >>>> experienced several compilation errors when trying to >>>> build the QuantExt library. >>>> >>>> I am using the official release of the 1.27.1 version >>>> of the QuantLib library as suggested in your email. >>>> >>>> One simple example of the several failures concerns the >>>> second CPICoupon constructor implemented online in the >>>> header cpicoupon.hpp >>>> >>>> Its top part looks as below: >>>> >>>> */CPICoupon(Real baseCPI, >>>> const Date& baseDate, // user provided, >>>> could be arbitrary >>>> const Date& paymentDate, Real nominal, >>>> const Date& startDate, const Date& endDate,/* >>>> >>>> */ (... more input arguments here ...)/* >>>> >>>> */ : QuantLib::CPICoupon(baseCPI, baseDate, >>>> paymentDate, nominal, startDate, endDate, index, >>>> observationLag,/* >>>> >>>> The compile-time error is caused by the implementation >>>> line: >>>> >>>> */QuantLib::CPICoupon(baseCPI, baseDate, paymentDate, >>>> nominal, startDate, endDate, index,/* >>>> >>>> The above line passes the argument */baseDate /*to the >>>> constructor of the */QuantLib::CPICoupon /*class. >>>> >>>> But this fails because the CPICoupon class in QuantLib >>>> does not contain any constructor that expects a second >>>> argument of type Date. >>>> >>>> Best regards, >>>> >>>> Ioannis Rigopoulos >>>> >>>> On 9/22/2022 5:29 PM, Roland Lichters via QuantLib-dev >>>> wrote: >>>> >>>> Dear all, >>>> >>>> we have just published the 7th release of ORE and >>>> ORE-SWIG, updating the codebase at >>>> https://github.com/OpenSourceRisk >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FOpenSourceRisk&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660004960075%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=E5qyW6bz6HQr%2F7oX8jjB6CapFKyvOT%2BGsG5yjmE8jgo%3D&reserved=0>. >>>> ORE 7 depends on QuantLib 1.27.1 and QuantLib-SWIG >>>> 1.27. >>>> >>>> The release contains many fixes and improvements >>>> over the past year that were triggered by our main >>>> sponsor’s (Acadia) use of ORE in their Initial >>>> Margin Risk Generator services. The full release >>>> notes can be found in the user guide under the >>>> DOCUMENTATION menu on https://opensourcerisk.org >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fopensourcerisk.org%2F&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660005116456%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=0NJEv%2BU4J7%2BpBN8A0GiK47iTduCXRaGVnc5fQZcoHlU%3D&reserved=0>. >>>> >>>> With this release we have started publishing a wide >>>> range of additional financial instruments. You will >>>> see various Equity/FX Exotics in this release. >>>> Commodity, Credit, Hybrids, Exotics with scripted >>>> payoffs will follow in the next releases in >>>> quarterly steps, see also the Acadia press release >>>> and roadmap here >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.acadia.inc%2Fnews%2Facadia-announces-seventh-release-of-open-source-risk-engine-with-quarterly-releases-to-follow&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660005116456%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=F0Uj0LGndGDHUqM573QS2VPbSGKfv%2FSCAz1nMYxJztM%3D&reserved=0>. >>>> >>>> >>>> If you want to hear from Acadia about latest ORE >>>> developments, then feel free to sign up here >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fshare.hsforms.com%2F1eqcUZ-9_QdSH__M7_YPSig43ul4&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660005116456%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=s5ye1RwJNE3OVVAPYeXyAaDyPI8iW7De8TE1lnVxuMI%3D&reserved=0> >>>> to receive updates by email. >>>> >>>> Please explore ORE – download the release >>>> executable, or clone the repositories and build. >>>> >>>> As usual, all feedback is welcome! >>>> >>>> Best regards, >>>> >>>> Roland >>>> >>>> >>>> /The information contained in this e-mail, and any >>>> attachment, is confidential and is intended solely >>>> for the use of the intended recipient. Access, >>>> copying or re-use of the e-mail or any attachment, >>>> or any information contained therein, by any other >>>> person is not authorized. If you are not the >>>> intended recipient please return the e-mail to the >>>> sender and delete it from your computer. The >>>> acadia.inc privacy policy is available on our >>>> website. / >>>> >>>> >>>> >>>> _______________________________________________ >>>> >>>> QuantLib-dev mailing list >>>> >>>> Qua...@li... >>>> >>>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.sourceforge.net%2Flists%2Flistinfo%2Fquantlib-dev&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660005116456%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=TqiZ79kCKWOAYyYp5j9dsrhEATGPZJ6iCvM7RnPu0mE%3D&reserved=0> >>>> >>>> width= >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.avast.com%2Fsig-email%3Futm_medium%3Demail%26utm_source%3Dlink%26utm_campaign%3Dsig-email%26utm_content%3Demailclient&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660005116456%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=wl6pCFww4PCCz7iUnlp2xPrNg6Wqa4MYzWZ5WCR1RaQ%3D&reserved=0> >>>> >>>> >>>> >>>> Virenfrei.www.avast.com >>>> <https://nam10.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.avast.com%2Fsig-email%3Futm_medium%3Demail%26utm_source%3Dlink%26utm_campaign%3Dsig-email%26utm_content%3Demailclient&data=05%7C01%7Croland.lichters%40acadia.inc%7C293729e77b074113491608daa5dba252%7Cf432debe2aae495fb2dfe8458cace1da%7C0%7C0%7C638004660005116456%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=wl6pCFww4PCCz7iUnlp2xPrNg6Wqa4MYzWZ5WCR1RaQ%3D&reserved=0> >>>> >>>> >>>> /The information contained in this e-mail, and any >>>> attachment, is confidential and is intended solely for >>>> the use of the intended recipient. Access, copying or >>>> re-use of the e-mail or any attachment, or any >>>> information contained therein, by any other person is >>>> not authorized. If you are not the intended recipient >>>> please return the e-mail to the sender and delete it >>>> from your computer. The acadia.inc privacy policy is >>>> available on our website. / >>>> >>>> >>>> The information contained in this e-mail, and any >>>> attachment, is confidential and is intended solely for the >>>> use of the intended recipient. Access, copying or re-use of >>>> the e-mail or any attachment, or any information contained >>>> therein, by any other person is not authorized. If you are >>>> not the intended recipient please return the e-mail to the >>>> sender and delete it from your computer. The acadia.inc >>>> privacy policy is available on our website. >>> >>> >>> _______________________________________________ >>> QuantLib-dev mailing list >>> Qua...@li... >>> https://lists.sourceforge.net/lists/listinfo/quantlib-dev >> _______________________________________________ >> QuantLib-dev mailing list >> Qua...@li... >> https://lists.sourceforge.net/lists/listinfo/quantlib-dev >> -- Diese E-Mail wurde von Avast-Antivirussoftware auf Viren geprüft. www.avast.com |
